Rankings

Values with different source types or measurement conditions (comparison groups) are ranked separately. Overall rankings aggregate only models with values for the same set of comparison groups, and costs are not combined across benchmarks. See About for the methods.

Ranks, percentiles, and frontiers include only models currently available in Claude Code / Codex (models: 8, excluding legacy models). For rankings including legacy and unavailable models, see All model rankings (records). Ranks differ because the set of models differs.

Overall (performance first)

Only model × effort entries with values for the same set of comparison groups are aggregated. Within each group, included entries are ranked and converted to percentiles (first = 100, last = 0), averaged across groups of the same benchmark, then averaged across benchmarks for ranking. Multiple groups for one benchmark do not give it extra weight. Aggregations cover different groups, so their scores are not comparable.

Aggregations are ordered by descending model count, benchmark count, then entry count. The aggregation containing the most models currently available in Claude Code / Codex opens first. The rest are collapsed under “Other aggregations” by source type.

Aggregation criteria: minimum models: 3 (different efforts count as one model); minimum benchmarks: 3 (multiple comparison groups for the same benchmark count as one). Coverage n/N counts benchmarks; N is the number of benchmarks with values of the same source type. “measured/reported by” for each group shows the parties and counts for values included in that aggregation. Even in official announcements, the publisher may have measured values for other companies’ models.

third-party · Models: 8 · Entries: 8 × Groups: 4 (benchmark coverage 4/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), SimpleQA Verified

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (8)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (8)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (8)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (8)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. SimpleQA Verified
1	Claude Fable 5.2	Max	84	6/10	93.9%	95.1%	100.0%	94.0%
1	GPT-6 Astra	Max	84	9/10	95.8%	93.7%	97.6%	75.6%
3	GPT-6.1 Sol	Max	79	9/10	95.4%	93.7%	100.0%	73.9%
4	Claude Opus 5.5	Max	50	7/10	90.6%	91.2%	95.0%	72.2%
5	GPT-6.1 Terra	Max	46	6/10	94.3%	90.9%	92.7%	57.3%
6	Claude Sonnet 5.5	Max	43	7/10	95.6%	88.8%	80.5%	46.5%
7	GPT-6 Luna	Max	14	9/10	90.5%	79.0%	56.1%	41.4%
8	Claude Haiku 5.5	Max	0	6/10	89.7%	75.1%	46.3%	23.8%

Other aggregations (entries: 11)

official

official · Models: 3 · Entries: 3 × Groups: 7 (benchmark coverage 7/27)FrontierSWE 2, AutomationBench, Terminal-Bench-Science 0.1, OSWorld 2.1 (offline set, partial), RoboDojo Sim, AA-Briefcase v1.1(rubric), AA-Briefcase v1.1(Elo)

FrontierSWE 2official · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (2) · not stated (1)
AutomationBenchofficial · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (2) · not stated (1)
Terminal-Bench-Science 0.1official · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (2) · not stated (1)
OSWorld 2.1 (offline set, partial)official · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (2) · not stated (1)
RoboDojo Simofficial · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (3)
AA-Briefcase v1.1(rubric)official · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (2) · not stated (1)
AA-Briefcase v1.1(Elo)official · Introducing Claude Fable 5.2 and Claude Mythos 5.2measured/reported by: Anthropic (2) · not stated (1)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. FrontierSWE 2	2. AutomationBench	3. Terminal-Bench-Science 0.1	4. OSWorld 2.1 (offline set, partial)	5. RoboDojo Sim	6. AA-Briefcase v1.1(rubric)	7. AA-Briefcase v1.1(Elo)
1	Claude Fable 5.2	not stated	93	7/27	69.9%	51.4%	74.6%	89.3%	64.9%	66.1%	1,785
2	Claude Opus 5.5	not stated	43	14/27	62.3%	42.5%	58.7%	87.2%	39.4%	61.2%	1,800
3	GPT-6 Astra	not stated	14	12/27	65.5%	41.4%	68.1%	79.5%	29.8%	52.0%	1,569

third-party

third-party · Models: 7 · Entries: 7 × Groups: 5 (benchmark coverage 5/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), OTIS Mock AIME 2024-2025, SimpleQA Verified

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (7)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (7)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (7)
OTIS Mock AIME 2024-2025third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0measured/reported by: Epoch AI (7)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (7)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. OTIS Mock AIME 2024-2025	5. SimpleQA Verified
1	GPT-6 Astra	Max	78	9/10	95.8%	93.7%	97.6%	100.0%	75.6%
2	GPT-6.1 Sol	Max	73	9/10	95.4%	93.7%	100.0%	100.0%	73.9%
3	Claude Fable 5.2	Max	68	6/10	93.9%	95.1%	100.0%	99.7%	94.0%
4	Claude Opus 5.5	Max	47	7/10	90.6%	91.2%	95.0%	100.0%	72.2%
5	GPT-6.1 Terra	Max	43	6/10	94.3%	90.9%	92.7%	100.0%	57.3%
6	Claude Sonnet 5.5	Max	40	7/10	95.6%	88.8%	80.5%	100.0%	46.5%
7	GPT-6 Luna	Max	0	9/10	90.5%	79.0%	56.1%	98.9%	41.4%

third-party · Models: 7 · Entries: 7 × Groups: 5 (benchmark coverage 5/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), SimpleQA Verified, Chess Puzzles(Epoch AI)

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (7)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (7)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (7)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (7)
Chess Puzzles(Epoch AI)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0measured/reported by: Epoch AI (7)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. SimpleQA Verified	5. Chess Puzzles(Epoch AI)
1	Claude Fable 5.2	Max	88	6/10	93.9%	95.1%	100.0%	94.0%	79.0%
2	GPT-6 Astra	Max	82	9/10	95.8%	93.7%	97.6%	75.6%	74.0%
3	GPT-6.1 Sol	Max	73	9/10	95.4%	93.7%	100.0%	73.9%	68.0%
4	Claude Opus 5.5	Max	47	7/10	90.6%	91.2%	95.0%	72.2%	67.0%
5	Claude Sonnet 5.5	Max	43	7/10	95.6%	88.8%	80.5%	46.5%	59.0%
6	GPT-6 Luna	Max	13	9/10	90.5%	79.0%	56.1%	41.4%	36.0%
7	Claude Haiku 5.5	Max	3	6/10	89.7%	75.1%	46.3%	23.8%	55.0%

third-party · Models: 7 · Entries: 7 × Groups: 5 (benchmark coverage 5/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), SimpleQA Verified, MirrorCode

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (7)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (7)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (7)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (7)
MirrorCodethird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0measured/reported by: Epoch AI (7)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. SimpleQA Verified	5. MirrorCode
1	GPT-6 Astra	Max	78	9/10	95.8%	93.7%	97.6%	75.6%	60.0%
2	GPT-6.1 Sol	Max	75	9/10	95.4%	93.7%	100.0%	73.9%	61.1%
3	Claude Opus 5.5	Max	63	7/10	90.6%	91.2%	95.0%	72.2%	95.6%
4	Claude Sonnet 5.5	Max	57	7/10	95.6%	88.8%	80.5%	46.5%	97.8%
5	GPT-6.1 Terra	Max	50	6/10	94.3%	90.9%	92.7%	57.3%	76.7%
6	Claude Haiku 5.5	Max	13	6/10	89.7%	75.1%	46.3%	23.8%	78.9%
6	GPT-6 Luna	Max	13	9/10	90.5%	79.0%	56.1%	41.4%	26.7%

third-party · Models: 6 · Entries: 6 × Groups: 6 (benchmark coverage 6/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), OTIS Mock AIME 2024-2025, SimpleQA Verified, Chess Puzzles(Epoch AI)

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (6)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (6)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (6)
OTIS Mock AIME 2024-2025third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0measured/reported by: Epoch AI (6)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (6)
Chess Puzzles(Epoch AI)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0measured/reported by: Epoch AI (6)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. OTIS Mock AIME 2024-2025	5. SimpleQA Verified	6. Chess Puzzles(Epoch AI)
1	GPT-6 Astra	Max	77	9/10	95.8%	93.7%	97.6%	100.0%	75.6%	74.0%
2	Claude Fable 5.2	Max	75	6/10	93.9%	95.1%	100.0%	99.7%	94.0%	79.0%
3	GPT-6.1 Sol	Max	68	9/10	95.4%	93.7%	100.0%	100.0%	73.9%	68.0%
4	Claude Opus 5.5	Max	42	7/10	90.6%	91.2%	95.0%	100.0%	72.2%	67.0%
5	Claude Sonnet 5.5	Max	38	7/10	95.6%	88.8%	80.5%	100.0%	46.5%	59.0%
6	GPT-6 Luna	Max	0	9/10	90.5%	79.0%	56.1%	98.9%	41.4%	36.0%

third-party · Models: 6 · Entries: 6 × Groups: 6 (benchmark coverage 6/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), OTIS Mock AIME 2024-2025, SimpleQA Verified, MirrorCode

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (6)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (6)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (6)
OTIS Mock AIME 2024-2025third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0measured/reported by: Epoch AI (6)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (6)
MirrorCodethird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0measured/reported by: Epoch AI (6)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. OTIS Mock AIME 2024-2025	5. SimpleQA Verified	6. MirrorCode
1	GPT-6 Astra	Max	75	9/10	95.8%	93.7%	97.6%	100.0%	75.6%	60.0%
2	GPT-6.1 Sol	Max	72	9/10	95.4%	93.7%	100.0%	100.0%	73.9%	61.1%
3	Claude Opus 5.5	Max	57	7/10	90.6%	91.2%	95.0%	100.0%	72.2%	95.6%
4	Claude Sonnet 5.5	Max	50	7/10	95.6%	88.8%	80.5%	100.0%	46.5%	97.8%
5	GPT-6.1 Terra	Max	47	6/10	94.3%	90.9%	92.7%	100.0%	57.3%	76.7%
6	GPT-6 Luna	Max	0	9/10	90.5%	79.0%	56.1%	98.9%	41.4%	26.7%

third-party · Models: 6 · Entries: 6 × Groups: 6 (benchmark coverage 6/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), SimpleQA Verified, Chess Puzzles(Epoch AI), MirrorCode

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (6)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (6)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (6)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (6)
Chess Puzzles(Epoch AI)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0measured/reported by: Epoch AI (6)
MirrorCodethird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0measured/reported by: Epoch AI (6)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. SimpleQA Verified	5. Chess Puzzles(Epoch AI)	6. MirrorCode
1	GPT-6 Astra	Max	82	9/10	95.8%	93.7%	97.6%	75.6%	74.0%	60.0%
2	GPT-6.1 Sol	Max	75	9/10	95.4%	93.7%	100.0%	73.9%	68.0%	61.1%
3	Claude Opus 5.5	Max	60	7/10	90.6%	91.2%	95.0%	72.2%	67.0%	95.6%
4	Claude Sonnet 5.5	Max	57	7/10	95.6%	88.8%	80.5%	46.5%	59.0%	97.8%
5	Claude Haiku 5.5	Max	13	6/10	89.7%	75.1%	46.3%	23.8%	55.0%	78.9%
5	GPT-6 Luna	Max	13	9/10	90.5%	79.0%	56.1%	41.4%	36.0%	26.7%

third-party · Models: 5 · Entries: 5 × Groups: 7 (benchmark coverage 7/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), OTIS Mock AIME 2024-2025, SimpleQA Verified, Chess Puzzles(Epoch AI), MirrorCode

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (5)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (5)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (5)
OTIS Mock AIME 2024-2025third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0measured/reported by: Epoch AI (5)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (5)
Chess Puzzles(Epoch AI)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0measured/reported by: Epoch AI (5)
MirrorCodethird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0measured/reported by: Epoch AI (5)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. OTIS Mock AIME 2024-2025	5. SimpleQA Verified	6. Chess Puzzles(Epoch AI)	7. MirrorCode
1	GPT-6 Astra	Max	79	9/10	95.8%	93.7%	97.6%	100.0%	75.6%	74.0%	60.0%
2	GPT-6.1 Sol	Max	71	9/10	95.4%	93.7%	100.0%	100.0%	73.9%	68.0%	61.1%
3	Claude Opus 5.5	Max	52	7/10	90.6%	91.2%	95.0%	100.0%	72.2%	67.0%	95.6%
4	Claude Sonnet 5.5	Max	48	7/10	95.6%	88.8%	80.5%	100.0%	46.5%	59.0%	97.8%
5	GPT-6 Luna	Max	0	9/10	90.5%	79.0%	56.1%	98.9%	41.4%	36.0%	26.7%

third-party · Models: 4 · Entries: 4 × Groups: 3 (benchmark coverage 3/10)OTIS Mock AIME 2024-2025, LMArena Text (style control, overall), LMArena Text Coding (without style control)

OTIS Mock AIME 2024-2025third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0measured/reported by: Epoch AI (4)
LMArena Text (style control, overall)third-party · lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)measured/reported by: LMArena (4)
LMArena Text Coding (without style control)third-party · lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)measured/reported by: LMArena (4)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. OTIS Mock AIME 2024-2025	2. LMArena Text (style control, overall)	3. LMArena Text Coding (without style control)
1	GPT-6.1 Sol	Max	83	9/10	100.0%	1,482	1,486
1	GPT-6 Astra	Max	83	9/10	100.0%	1,474	1,494
3	GPT-6 Luna	Max	22	9/10	98.9%	1,441	1,432
4	Claude Haiku 5.5	High	11	3/10	97.2%	1,422	1,454

third-party · Models: 4 · Entries: 4 × Groups: 3 (benchmark coverage 3/10)MirrorCode, LMArena Text (style control, overall), LMArena Text Coding (without style control)

MirrorCodethird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0measured/reported by: Epoch AI (4)
LMArena Text (style control, overall)third-party · lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)measured/reported by: LMArena (4)
LMArena Text Coding (without style control)third-party · lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)measured/reported by: LMArena (4)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. MirrorCode	2. LMArena Text (style control, overall)	3. LMArena Text Coding (without style control)
1	Claude Opus 5.5	High	100	3/10	78.9%	1,506	1,542
2	GPT-6.1 Sol	Max	56	9/10	61.1%	1,482	1,486
3	GPT-6 Astra	Max	44	9/10	60.0%	1,474	1,494
4	GPT-6 Luna	Max	0	9/10	26.7%	1,441	1,432

third-party · Models: 3 · Entries: 3 × Groups: 9 (benchmark coverage 9/10)GPQA Diamond, FrontierMath Tiers 1-3(v2), FrontierMath Tier 4(v2), OTIS Mock AIME 2024-2025, SimpleQA Verified, Chess Puzzles(Epoch AI), MirrorCode, LMArena Text (style control, overall), LMArena Text Coding (without style control)

GPQA Diamondthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0measured/reported by: Epoch AI (3)
FrontierMath Tiers 1-3(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0measured/reported by: Epoch AI (3)
FrontierMath Tier 4(v2)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0measured/reported by: Epoch AI (3)
OTIS Mock AIME 2024-2025third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0measured/reported by: Epoch AI (3)
SimpleQA Verifiedthird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0measured/reported by: Epoch AI (3)
Chess Puzzles(Epoch AI)third-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0measured/reported by: Epoch AI (3)
MirrorCodethird-party · Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0measured/reported by: Epoch AI (3)
LMArena Text (style control, overall)third-party · lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)measured/reported by: LMArena (3)
LMArena Text Coding (without style control)third-party · lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)measured/reported by: LMArena (3)

Rank	Model	effort	Mean percentile	Benchmarks with values	1. GPQA Diamond	2. FrontierMath Tiers 1-3(v2)	3. FrontierMath Tier 4(v2)	4. OTIS Mock AIME 2024-2025	5. SimpleQA Verified	6. Chess Puzzles(Epoch AI)	7. MirrorCode	8. LMArena Text (style control, overall)	9. LMArena Text Coding (without style control)
1	GPT-6 Astra	Max	78	9/10	95.8%	93.7%	97.6%	100.0%	75.6%	74.0%	60.0%	1,474	1,494
2	GPT-6.1 Sol	Max	72	9/10	95.4%	93.7%	100.0%	100.0%	73.9%	68.0%	61.1%	1,482	1,486
3	GPT-6 Luna	Max	0	9/10	90.5%	79.0%	56.1%	98.9%	41.4%	36.0%	26.7%	1,441	1,432

Cost balance

Source costs have different meanings across benchmarks, so they are considered only within comparison groups and never combined across benchmarks.

Frontier: Points for which no better score exists at the same or lower cost (Pareto frontier), ordered by cost. Points without source costs are excluded.
Effort changes: Score differences (in points for percentage metrics) and cost ratios between adjacent efforts with values for the same model. Missing intermediate efforts are noted. When measured/reported by differs at the two ends, it is shown as “before → after”.

Terminal-Bench 4.0

official results (Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Sonnet 5.5	Low	20.0%	$0.76Cost per attempt	Anthropic
Claude Sonnet 5.5	Med	28.8%	$0.83Cost per attempt	Anthropic
Claude Opus 5.5	Low	38.5%	$1.29Cost per attempt	Anthropic
Claude Sonnet 5.5	High	43.0%	$1.94Cost per attempt	Anthropic
Claude Opus 5.5	Med	57.6%	$2.94Cost per attempt	Anthropic
Claude Opus 5.5	High	64.2%	$3.88Cost per attempt	Anthropic
Claude Opus 5.5	Xhigh	66.4%	$7.35Cost per attempt	Anthropic
Claude Sonnet 5.5	Max	70.6%	$12.54Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+19.1pt	×2.28	Anthropic
Claude Opus 5.5	Med → High	+6.6pt	×1.32	Anthropic
Claude Opus 5.5	High → Xhigh	+2.2pt	×1.89	Anthropic
Claude Opus 5.5	Xhigh → Max	-1.6pt	×1.53	Anthropic
Claude Sonnet 5.5	Low → Med	+8.8pt	×1.09	Anthropic
Claude Sonnet 5.5	Med → High	+14.2pt	×2.34	Anthropic
Claude Sonnet 5.5	High → Xhigh	+18.5pt	×2.73	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+9.1pt	×2.37	Anthropic
GPT-6 Astra	Low → Med	+4.2pt	×1.24	OpenAI
GPT-6 Astra	Med → High	+4.0pt	×1.17	OpenAI
GPT-6 Astra	High → Xhigh	-0.3pt	×1.04	OpenAI
GPT-6 Astra	Xhigh → Max	-0.9pt	×1.38	OpenAI

official results (Introducing Claude Haiku 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Haiku 5.5	Low	12.7%	$0.42Cost per attempt	Anthropic
Claude Sonnet 5.5	Low	20.0%	$0.62Cost per attempt	Anthropic
Claude Sonnet 5.5	Med	28.8%	$0.68Cost per attempt	Anthropic
Claude Sonnet 5.5	High	43.0%	$1.46Cost per attempt	Anthropic
Claude Sonnet 5.5	Xhigh	61.5%	$4.34Cost per attempt	Anthropic
Claude Sonnet 5.5	Max	70.6%	$10.44Cost per attempt	Anthropic

Points without source costs are excluded from the frontier (excluded: 1).

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Sonnet 5.5	Low → Med	+8.8pt	×1.1	Anthropic
Claude Sonnet 5.5	Med → High	+14.2pt	×2.15	Anthropic
Claude Sonnet 5.5	High → Xhigh	+18.5pt	×2.97	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+9.1pt	×2.41	Anthropic
Claude Haiku 5.5	Low → Med	+7.6pt	×1.62	Anthropic
Claude Haiku 5.5	Med → High	+4.5pt	×1.53	Anthropic
Claude Haiku 5.5	High → Xhigh	+6.7pt	×1.68	Anthropic
Claude Haiku 5.5	Xhigh → Max	+7.7pt	×1.51	Anthropic

FrontierCode v1.1(main set)

official results (Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Sonnet 5.5	Low	29.3%	$0.19Cost per task	Anthropic
Claude Sonnet 5.5	Med	36.5%	$0.24Cost per task	Anthropic
Claude Opus 5.5	Low	47.3%	$0.40Cost per task	Anthropic
Claude Sonnet 5.5	High	49.4%	$0.42Cost per task	Anthropic
Claude Opus 5.5	Med	54.6%	$0.80Cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+7.3pt	×2.0	Anthropic
Claude Opus 5.5	Med → High	-0.6pt	×1.36	Anthropic
Claude Opus 5.5	High → Xhigh	-2.6pt	×2.06	Anthropic
Claude Opus 5.5	Xhigh → Max	+3.0pt	×2.75	Anthropic
Claude Sonnet 5.5	Low → Med	+7.2pt	×1.26	Anthropic
Claude Sonnet 5.5	Med → High	+12.9pt	×1.75	Anthropic
Claude Sonnet 5.5	High → Xhigh	+2.7pt	×3.79	Anthropic
Claude Sonnet 5.5	Xhigh → Max	-5.9pt	×13.07	Anthropic
GPT-6 Astra	Low → Med	+3.5pt	×1.43	not stated
GPT-6 Astra	Med → High	+2.1pt	×1.25	not stated
GPT-6 Astra	High → Xhigh	-0.3pt	×1.09	not stated
GPT-6 Astra	Xhigh → Max	+2.7pt	×1.41	not stated

FrontierCode 2.0

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	14.7%	$0.75Cost per task	Anthropic
Claude Opus 5.5	Med	30.1%	$1.70Cost per task	Anthropic
Claude Fable 5.2	Low	34.2%	$2.10Cost per task	Anthropic
Claude Opus 5.5	High	34.5%	$2.42Cost per task	Anthropic
Claude Fable 5.2	Med	37.6%	$2.48Cost per task	Anthropic
GPT-6 Astra	High	38.3%	$4.28Cost per task	not stated
GPT-6 Astra	Xhigh	41.4%	$4.54Cost per task	not stated
Claude Opus 5.5	Xhigh	43.4%	$5.18Cost per task	Anthropic
Claude Fable 5.2	High	45.3%	$6.72Cost per task	Anthropic
Claude Opus 5.5	Max	45.5%	$12.82Cost per task	Anthropic
Claude Fable 5.2	Xhigh	49.1%	$14.19Cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Low → Med	+3.4pt	×1.18	Anthropic
Claude Fable 5.2	Med → High	+7.7pt	×2.71	Anthropic
Claude Fable 5.2	High → Xhigh	+3.8pt	×2.11	Anthropic
Claude Fable 5.2	Xhigh → Max	-3.5pt	×3.15	Anthropic
Claude Opus 5.5	Low → Med	+15.4pt	×2.27	Anthropic
Claude Opus 5.5	Med → High	+4.4pt	×1.42	Anthropic
Claude Opus 5.5	High → Xhigh	+8.9pt	×2.14	Anthropic
Claude Opus 5.5	Xhigh → Max	+2.1pt	×2.47	Anthropic
GPT-6 Astra	Low → Med	+5.7pt	×1.43	not stated
GPT-6 Astra	Med → High	+1.5pt	×1.23	not stated
GPT-6 Astra	High → Xhigh	+3.1pt	×1.06	not stated
GPT-6 Astra	Xhigh → Max	+1.4pt	×1.47	not stated

CursorBench 4.0

official results (Introducing Claude Opus 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	43.7%	$1.18Cost per task	Anthropic
Claude Opus 5.5	Med	52.5%	$2.90Cost per task	Anthropic
Claude Opus 5.5	High	56.0%	$3.97Cost per task	Anthropic
Claude Opus 5.5	Xhigh	56.0%	$6.99Cost per task	Anthropic
Claude Opus 5.5	Max	57.8%	$13.43Cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+8.8pt	×2.46	Anthropic
Claude Opus 5.5	Med → High	+3.5pt	×1.37	Anthropic
Claude Opus 5.5	High → Xhigh	±0.0pt	×1.76	Anthropic
Claude Opus 5.5	Xhigh → Max	+1.8pt	×1.92	Anthropic

official results (Introducing Claude Sonnet 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Sonnet 5.5	Low	35.8%	$0.50Cost per task	Anthropic
Claude Sonnet 5.5	Med	39.2%	$0.70Cost per task	Anthropic
Claude Opus 5.5	Low	43.7%	$1.17Cost per task	Anthropic
Claude Sonnet 5.5	High	47.8%	$1.67Cost per task	Anthropic
Claude Opus 5.5	Med	52.5%	$2.91Cost per task	Anthropic
Claude Sonnet 5.5	Xhigh	53.1%	$3.88Cost per task	Anthropic
Claude Opus 5.5	High	56.0%	$3.97Cost per task	Anthropic
Claude Opus 5.5	Xhigh	56.0%	$6.98Cost per task	Anthropic
Claude Opus 5.5	Max	57.8%	$13.43Cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+8.8pt	×2.49	Anthropic
Claude Opus 5.5	Med → High	+3.5pt	×1.36	Anthropic
Claude Opus 5.5	High → Xhigh	±0.0pt	×1.76	Anthropic
Claude Opus 5.5	Xhigh → Max	+1.8pt	×1.92	Anthropic
Claude Sonnet 5.5	Low → Med	+3.4pt	×1.4	Anthropic
Claude Sonnet 5.5	Med → High	+8.6pt	×2.39	Anthropic
Claude Sonnet 5.5	High → Xhigh	+5.3pt	×2.32	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+2.4pt	×2.49	Anthropic

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

No frontier can be shown because no points have source costs.

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Low → Med	+1.6pt	—	Cursor
Claude Fable 5.2	Med → High	+2.1pt	—	Cursor
Claude Fable 5.2	High → Xhigh	-0.6pt	—	Cursor
Claude Fable 5.2	Xhigh → Max	+3.7pt	—	Cursor

GDPval-AA v2.1

official results (Introducing Claude Opus 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	1,224	$0.21Estimated cost per task	Anthropic
GPT-6 Astra	Low	1,366	$0.85Estimated cost per task	not stated
Claude Opus 5.5	Med	1,576	$0.86Estimated cost per task	Anthropic
Claude Opus 5.5	High	1,692	$1.54Estimated cost per task	Anthropic
Claude Opus 5.5	Xhigh	1,820	$4.21Estimated cost per task	Anthropic
Claude Opus 5.5	Max	1,846	$8.92Estimated cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+352	×4.1	Anthropic
Claude Opus 5.5	Med → High	+116	×1.79	Anthropic
Claude Opus 5.5	High → Xhigh	+128	×2.73	Anthropic
Claude Opus 5.5	Xhigh → Max	+26	×2.12	Anthropic
GPT-6 Astra	Low → Med	+102	×2.14	not stated
GPT-6 Astra	Med → High	+17	×1.34	not stated
GPT-6 Astra	High → Xhigh	+31	×1.25	not stated
GPT-6 Astra	Xhigh → Max	+26	×1.49	not stated

official results (Introducing Claude Haiku 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
GPT-6 Luna	Low	1,036	$0.004Cost per task	not stated
Claude Haiku 5.5	Low	1,125	$0.01Cost per task	Anthropic
GPT-6 Luna	Med	1,262	$0.02Cost per task	not stated
GPT-6 Luna	High	1,344	$0.03Cost per task	not stated
GPT-6 Luna	Xhigh	1,364	$0.05Cost per task	not stated
GPT-6 Luna	Max	1,437	$0.09Cost per task	not stated
Claude Haiku 5.5	Xhigh	1,513	$0.27Cost per task	Anthropic
Claude Sonnet 5.5	High	1,551	$0.62Cost per task	Anthropic
Claude Haiku 5.5	Max	1,620	$0.87Cost per task	Anthropic
Claude Sonnet 5.5	Xhigh	1,731	$1.88Cost per task	Anthropic
Claude Sonnet 5.5	Max	1,840	$6.78Cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Sonnet 5.5	Low → Med	+145	×1.23	Anthropic
Claude Sonnet 5.5	Med → High	+227	×2.3	Anthropic
Claude Sonnet 5.5	High → Xhigh	+180	×3.03	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+109	×3.61	Anthropic
Claude Haiku 5.5	Low → Med	+152	×3.0	Anthropic
Claude Haiku 5.5	Med → High	+143	×3.0	Anthropic
Claude Haiku 5.5	High → Xhigh	+93	×3.0	Anthropic
Claude Haiku 5.5	Xhigh → Max	+107	×3.22	Anthropic
GPT-6 Luna	Low → Med	+226	×5.0	not stated
GPT-6 Luna	Med → High	+82	×1.5	not stated
GPT-6 Luna	High → Xhigh	+20	×1.67	not stated
GPT-6 Luna	Xhigh → Max	+73	×1.8	not stated

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

No frontier can be shown because no points have source costs.

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Xhigh → Max	+80	—	Artificial Analysis

AA-Briefcase v1.1(Elo)

official results (Introducing Claude Sonnet 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Sonnet 5.5	Low	1,264	$0.87Cost per task	Anthropic
Claude Opus 5.5	Low	1,285	$1.15Cost per task	Anthropic
Claude Sonnet 5.5	Med	1,461	$1.64Cost per task	Anthropic
Claude Sonnet 5.5	High	1,634	$3.95Cost per task	Anthropic
Claude Opus 5.5	Med	1,642	$4.40Cost per task	Anthropic
Claude Opus 5.5	High	1,705	$6.27Cost per task	Anthropic
Claude Sonnet 5.5	Xhigh	1,746	$9.63Cost per task	Anthropic
Claude Opus 5.5	Xhigh	1,780	$12.27Cost per task	Anthropic
Claude Opus 5.5	Max	1,822	$21.05Cost per task	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+357	×3.83	Anthropic
Claude Opus 5.5	Med → High	+63	×1.42	Anthropic
Claude Opus 5.5	High → Xhigh	+75	×1.96	Anthropic
Claude Opus 5.5	Xhigh → Max	+42	×1.72	Anthropic
Claude Sonnet 5.5	Low → Med	+197	×1.89	Anthropic
Claude Sonnet 5.5	Med → High	+173	×2.41	Anthropic
Claude Sonnet 5.5	High → Xhigh	+112	×2.44	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+65	×3.03	Anthropic

AutomationBench

official results (Introducing Claude Opus 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	23.3%	$0.50Cost per task	Zapier
Claude Opus 5.5	Med	28.6%	$0.64Cost per task	Zapier
Claude Opus 5.5	High	32.0%	$0.70Cost per task	Zapier
Claude Opus 5.5	Xhigh	34.4%	$0.86Cost per task	Zapier
Claude Opus 5.5	Max	40.0%	$1.37Cost per task	Zapier
GPT-6 Astra	Max	41.4%	$1.77Cost per task	Zapier

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+5.3pt	×1.28	Zapier
Claude Opus 5.5	Med → High	+3.4pt	×1.09	Zapier
Claude Opus 5.5	High → Xhigh	+2.4pt	×1.23	Zapier
Claude Opus 5.5	Xhigh → Max	+5.6pt	×1.59	Zapier
GPT-6 Astra	Low → Med	+3.8pt	×1.19	Zapier
GPT-6 Astra	Med → High	+3.0pt	×1.13	Zapier
GPT-6 Astra	High → Xhigh	+1.9pt	×1.06	Zapier
GPT-6 Astra	Xhigh → Max	+2.4pt	×1.16	Zapier

Humanity's Last Exam (with tools)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	57.4%	$0.08Cost per attempt	Anthropic
Claude Opus 5.5	Med	63.0%	$0.17Cost per attempt	Anthropic
Claude Fable 5.2	Low	68.4%	$0.19Cost per attempt	Anthropic
Claude Fable 5.2	Med	68.6%	$0.21Cost per attempt	Anthropic
Claude Fable 5.2	High	71.5%	$0.63Cost per attempt	Anthropic
Claude Fable 5.2	Xhigh	73.4%	$1.48Cost per attempt	Anthropic
Claude Fable 5.2	Max	74.2%	$4.51Cost per attempt	Anthropic

Points without source costs are excluded from the frontier (excluded: 1).

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Low → Med	+0.2pt	×1.11	Anthropic
Claude Fable 5.2	Med → High	+2.9pt	×3.0	Anthropic
Claude Fable 5.2	High → Xhigh	+1.9pt	×2.35	Anthropic
Claude Fable 5.2	Xhigh → Max	+0.8pt	×3.05	Anthropic
Claude Opus 5.5	Low → Med	+5.6pt	×2.13	Anthropic
Claude Opus 5.5	Med → High	+0.9pt	×1.71	Anthropic
Claude Opus 5.5	High → Xhigh	+2.5pt	×2.38	Anthropic
Claude Opus 5.5	Xhigh → Max	+1.3pt	×3.16	Anthropic

Humanity's Last Exam (without tools)

official results (Introducing Claude Haiku 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Haiku 5.5	Low	30.9%	$0.003Cost per attempt	Anthropic
Claude Haiku 5.5	Med	35.5%	$0.007Cost per attempt	Anthropic
Claude Haiku 5.5	High	39.7%	$0.02Cost per attempt	Anthropic
Claude Sonnet 5.5	Med	43.2%	$0.03Cost per attempt	Anthropic
Claude Haiku 5.5	Xhigh	44.1%	$0.06Cost per attempt	Anthropic
Claude Sonnet 5.5	High	47.9%	$0.08Cost per attempt	Anthropic
Claude Sonnet 5.5	Xhigh	53.0%	$0.20Cost per attempt	Anthropic
Claude Sonnet 5.5	Max	56.9%	$0.89Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Sonnet 5.5	Low → Med	+1.8pt	×1.0	Anthropic
Claude Sonnet 5.5	Med → High	+4.7pt	×2.67	Anthropic
Claude Sonnet 5.5	High → Xhigh	+5.1pt	×2.5	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+3.9pt	×4.45	Anthropic
Claude Haiku 5.5	Low → Med	+4.6pt	×2.33	Anthropic
Claude Haiku 5.5	Med → High	+4.2pt	×2.86	Anthropic
Claude Haiku 5.5	High → Xhigh	+4.4pt	×3.0	Anthropic
Claude Haiku 5.5	Xhigh → Max	+1.8pt	×3.5	Anthropic

OSWorld 2.1(offline subset)

official results (Introducing Claude Haiku 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
GPT-6 Luna	Low	19.2%	$0.04Cost per attempt	not stated
Claude Haiku 5.5	Low	42.0%	$0.07Cost per attempt	Anthropic
Claude Haiku 5.5	Med	53.3%	$0.13Cost per attempt	Anthropic
Claude Haiku 5.5	High	61.3%	$0.18Cost per attempt	Anthropic
Claude Haiku 5.5	Xhigh	67.6%	$0.28Cost per attempt	Anthropic
Claude Haiku 5.5	Max	72.4%	$0.61Cost per attempt	Anthropic
Claude Sonnet 5.5	High	73.2%	$1.38Cost per attempt	Anthropic
Claude Sonnet 5.5	Xhigh	81.1%	$2.22Cost per attempt	Anthropic
Claude Sonnet 5.5	Max	83.9%	$5.73Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Sonnet 5.5	Low → Med	+8.1pt	×1.37	Anthropic
Claude Sonnet 5.5	Med → High	+7.2pt	×1.48	Anthropic
Claude Sonnet 5.5	High → Xhigh	+7.9pt	×1.61	Anthropic
Claude Sonnet 5.5	Xhigh → Max	+2.8pt	×2.58	Anthropic
Claude Haiku 5.5	Low → Med	+11.3pt	×1.86	Anthropic
Claude Haiku 5.5	Med → High	+8.0pt	×1.38	Anthropic
Claude Haiku 5.5	High → Xhigh	+6.3pt	×1.56	Anthropic
Claude Haiku 5.5	Xhigh → Max	+4.8pt	×2.18	Anthropic
GPT-6 Luna	Low → Med	+18.3pt	×3.25	not stated
GPT-6 Luna	Med → High	+4.8pt	×1.08	not stated
GPT-6 Luna	High → Xhigh	+2.5pt	×1.21	not stated
GPT-6 Luna	Xhigh → Max	+4.1pt	×1.18	not stated

Chartography (without tools)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	57.0%	$0.02Cost per attempt	Anthropic
Claude Fable 5.2	Med	66.4%	$0.03Cost per attempt	Anthropic
Claude Fable 5.2	High	69.8%	$0.07Cost per attempt	Anthropic
Claude Fable 5.2	Xhigh	71.4%	$0.15Cost per attempt	Anthropic
Claude Fable 5.2	Max	73.4%	$0.75Cost per attempt	Anthropic

Points without source costs are excluded from the frontier (excluded: 1).

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Low → Med	+2.2pt	×1.0	Anthropic
Claude Fable 5.2	Med → High	+3.4pt	×2.33	Anthropic
Claude Fable 5.2	High → Xhigh	+1.6pt	×2.14	Anthropic
Claude Fable 5.2	Xhigh → Max	+2.0pt	×5.0	Anthropic
Claude Opus 5.5	Low → Med	+2.0pt	×1.5	Anthropic
Claude Opus 5.5	Med → High	+0.2pt	×1.0	Anthropic
Claude Opus 5.5	High → Xhigh	+3.2pt	×2.67	Anthropic
Claude Opus 5.5	Xhigh → Max	+2.0pt	×6.13	Anthropic

WANDR

official results (Introducing Claude Opus 5.5)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	31.2%	$1.20Cost per attempt	Anthropic
Claude Opus 5.5	Med	62.8%	$11.20Cost per attempt	Anthropic
Claude Opus 5.5	High	67.3%	$16.88Cost per attempt	Anthropic
Claude Opus 5.5	Xhigh	71.3%	$29.06Cost per attempt	Anthropic
Claude Opus 5.5	Max	72.3%	$37.92Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+31.6pt	×9.33	Anthropic
Claude Opus 5.5	Med → High	+4.5pt	×1.51	Anthropic
Claude Opus 5.5	High → Xhigh	+4.0pt	×1.72	Anthropic
Claude Opus 5.5	Xhigh → Max	+1.0pt	×1.3	Anthropic

WANDR(relaxed score)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	31.2%	$1.25Cost per attempt	Anthropic
Claude Opus 5.5	Med	62.8%	$11.40Cost per attempt	Anthropic
Claude Fable 5.2	Low	71.4%	$12.66Cost per attempt	Anthropic
Claude Fable 5.2	Med	73.6%	$15.71Cost per attempt	Anthropic
Claude Fable 5.2	High	79.3%	$29.20Cost per attempt	Anthropic
Claude Fable 5.2	Xhigh	80.6%	$39.84Cost per attempt	Anthropic
Claude Fable 5.2	Max	82.1%	$53.11Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Low → Med	+2.2pt	×1.24	Anthropic
Claude Fable 5.2	Med → High	+5.7pt	×1.86	Anthropic
Claude Fable 5.2	High → Xhigh	+1.3pt	×1.36	Anthropic
Claude Fable 5.2	Xhigh → Max	+1.5pt	×1.33	Anthropic
Claude Opus 5.5	Low → Med	+31.6pt	×9.12	Anthropic
Claude Opus 5.5	Med → High	+4.5pt	×1.52	Anthropic
Claude Opus 5.5	High → Xhigh	+4.0pt	×1.69	Anthropic
Claude Opus 5.5	Xhigh → Max	+1.0pt	×1.32	Anthropic

BenchCAD 2.1 (with tools)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	6.3	$0.03Cost per attempt	Anthropic
Claude Opus 5.5	Med	19.1	$1.11Cost per attempt	Anthropic
Claude Fable 5.2	Low	26.9	$1.89Cost per attempt	Anthropic
Claude Opus 5.5	High	29.7	$2.57Cost per attempt	Anthropic
Claude Fable 5.2	Med	32.4	$2.66Cost per attempt	Anthropic
Claude Opus 5.5	Xhigh	54.5	$8.91Cost per attempt	Anthropic
Claude Fable 5.2	High	61	$9.99Cost per attempt	Anthropic
Claude Opus 5.5	Max	65.3	$19.86Cost per attempt	Anthropic
Claude Fable 5.2	Xhigh	74.1	$23.23Cost per attempt	Anthropic
Claude Fable 5.2	Max	83.5	$50.77Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Fable 5.2	Low → Med	+5.5	×1.41	Anthropic
Claude Fable 5.2	Med → High	+28.6	×3.76	Anthropic
Claude Fable 5.2	High → Xhigh	+13.1	×2.33	Anthropic
Claude Fable 5.2	Xhigh → Max	+9.4	×2.19	Anthropic
Claude Opus 5.5	Low → Med	+12.8	×37.0	Anthropic
Claude Opus 5.5	Med → High	+10.6	×2.32	Anthropic
Claude Opus 5.5	High → Xhigh	+24.8	×3.47	Anthropic
Claude Opus 5.5	Xhigh → Max	+10.8	×2.23	Anthropic

CVE Rediscovery

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Points on the frontier (lowest cost first)
Models on the frontier	effort	Score	Source cost (metric)	measured/reported by
Claude Opus 5.5	Low	59.4%	$0.11Cost per attempt	Anthropic
Claude Opus 5.5	Med	66.1%	$0.28Cost per attempt	Anthropic
Claude Opus 5.5	High	70.0%	$0.43Cost per attempt	Anthropic
Claude Opus 5.5	Xhigh	76.4%	$1.05Cost per attempt	Anthropic
Claude Opus 5.5	Max	81.3%	$2.98Cost per attempt	Anthropic

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	Low → Med	+6.7pt	×2.55	Anthropic
Claude Opus 5.5	Med → High	+3.9pt	×1.54	Anthropic
Claude Opus 5.5	High → Xhigh	+6.4pt	×2.44	Anthropic
Claude Opus 5.5	Xhigh → Max	+4.9pt	×2.84	Anthropic

FrontierMath Tier 4(v2)

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0)

No frontier can be shown because no points have source costs.

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
GPT-6 Astra	Low → Med	+9.8pt	—	Epoch AI
GPT-6 Astra	Med → High	±0.0pt	—	Epoch AI
GPT-6 Astra	High → Xhigh	±0.0pt	—	Epoch AI
GPT-6 Astra	Xhigh → Max	±0.0pt	—	Epoch AI

OTIS Mock AIME 2024-2025

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0)

No frontier can be shown because no points have source costs.

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Haiku 5.5	High → Xhigh	+1.7pt	—	Epoch AI

Chess Puzzles(Epoch AI)

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0)

No frontier can be shown because no points have source costs.

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Haiku 5.5	Low → MaxNo value for Med · High · Xhigh	+31.0pt	—	Epoch AI

MirrorCode

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0)

No frontier can be shown because no points have source costs.

Changes with increasing effort
Model	effort	Score difference	Cost ratio	measured/reported by
Claude Opus 5.5	High → MaxNo value for Xhigh	+16.7pt	—	Epoch AI
GPT-6 Astra	High → MaxNo value for Xhigh	+13.3pt	—	Epoch AI

By metric

Benchmarks are ordered by category and ranked within each comparison group (the same source, or sources verified to use the same measurement conditions). “measured/reported by” identifies the party stated in the source as measuring or reporting the value. Even in official announcements, the publisher may have measured values for other companies’ models.

Agentic codingTerminal-Bench 4.0

official results (Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	70.6%	$12.54Cost per attempt	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Xhigh	66.4%	$7.35Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
3	Claude Opus 5.5	Max	64.8%	$11.24Cost per attempt	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
4	Claude Opus 5.5	High	64.2%	$3.88Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
5	Claude Sonnet 5.5	Xhigh	61.5%	$5.30Cost per attempt	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
6	GPT-6 Astra	High	57.9%	$7.21Cost per attempt	—	OpenAI	Introducing Claude Opus 5.5Retrieved 2026-11-12
7	Claude Opus 5.5	Med	57.6%	$2.94Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
7	GPT-6 Astra	Xhigh	57.6%	$7.48Cost per attempt	—	OpenAI	Introducing Claude Opus 5.5Retrieved 2026-11-12
9	GPT-6 Astra	Max	56.7%	$10.35Cost per attempt	—	OpenAI	Introducing Claude Opus 5.5Retrieved 2026-11-12
10	GPT-6 Astra	Med	53.9%	$6.15Cost per attempt	—	OpenAI	Introducing Claude Opus 5.5Retrieved 2026-11-12
11	GPT-6 Astra	Low	49.7%	$4.95Cost per attempt	—	OpenAI	Introducing Claude Opus 5.5Retrieved 2026-11-12
12	Claude Sonnet 5.5	High	43.0%	$1.94Cost per attempt	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
13	Claude Opus 5.5	Low	38.5%	$1.29Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
14	Claude Sonnet 5.5	Med	28.8%	$0.83Cost per attempt	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
15	Claude Sonnet 5.5	Low	20.0%	$0.76Cost per attempt	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	70.6%	$10.44Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	Xhigh	61.5%	$4.34Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	Claude Sonnet 5.5	High	43.0%	$1.46Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
4	Claude Haiku 5.5	Max	39.2%	$2.64Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
5	Claude Haiku 5.5	Xhigh	31.5%	$1.75Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
6	Claude Sonnet 5.5	Med	28.8%	$0.68Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
7	Claude Haiku 5.5	High	24.8%	$1.04Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
8	Claude Haiku 5.5	Med	20.3%	$0.68Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
9	Claude Sonnet 5.5	Low	20.0%	$0.62Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
10	GPT-6 Luna	not stated	16.4%	—	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
11	Claude Haiku 5.5	Low	12.7%	$0.42Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	67.6%	—	—	Anthropic	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

Agentic codingTerminal-Bench-Science 0.1

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6 Astra	not stated	64.6%	—	—	OpenAI	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Max	58.7%	—	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	74.6%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	GPT-6 Astra	not stated	68.1%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Opus 5.5	not stated	58.7%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

official results (Claude Sonnet 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	59.9%	—	—	Anthropic	Claude Sonnet 5.5 System CardRetrieved 2026-11-14

official results (Claude Haiku 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6.1 Sol	Max	57.0%	—	—	OpenAI	Claude Haiku 5.5 System CardRetrieved 2026-11-14
2	Claude Haiku 5.5	Max	20.6%	—	—	Anthropic	Claude Haiku 5.5 System CardRetrieved 2026-11-14

Agentic codingFrontierCode v1.1(main set)

official results (Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Med	54.6%	$0.80Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Max	54.4%	$6.19Cost per task	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
3	Claude Opus 5.5	High	54.0%	$1.09Cost per task	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
4	GPT-6 Astra	Max	53.3%	$4.36Cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
5	Claude Sonnet 5.5	Xhigh	52.1%	$1.59Cost per task	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
6	Claude Opus 5.5	Xhigh	51.4%	$2.25Cost per task	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
7	GPT-6 Astra	High	50.9%	$2.85Cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
8	GPT-6 Astra	Xhigh	50.6%	$3.10Cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
9	Claude Sonnet 5.5	High	49.4%	$0.42Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
10	GPT-6 Astra	Med	48.8%	$2.28Cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
11	Claude Opus 5.5	Low	47.3%	$0.40Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
12	Claude Sonnet 5.5	Max	46.2%	$20.78Cost per task	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
13	GPT-6 Astra	Low	45.3%	$1.59Cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
14	Claude Sonnet 5.5	Med	36.5%	$0.24Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
15	Claude Sonnet 5.5	Low	29.3%	$0.19Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Xhigh	52.1%	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Haiku 5.5	not stated	46.4%	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	GPT-6 Luna	not stated	42.4%	—	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12

Agentic codingFrontierCode 2.0

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Xhigh	49.1%	$14.19Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Fable 5.2	Max	45.6%	$44.70Cost per task	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Opus 5.5	Max	45.5%	$12.82Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
4	Claude Fable 5.2	High	45.3%	$6.72Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
5	Claude Opus 5.5	Xhigh	43.4%	$5.18Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
6	GPT-6 Astra	Max	42.8%	$6.68Cost per task	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
7	GPT-6 Astra	Xhigh	41.4%	$4.54Cost per task	frontier	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
8	GPT-6 Astra	High	38.3%	$4.28Cost per task	frontier	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
9	Claude Fable 5.2	Med	37.6%	$2.48Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
10	GPT-6 Astra	Med	36.8%	$3.49Cost per task	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
11	Claude Opus 5.5	High	34.5%	$2.42Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
12	Claude Fable 5.2	Low	34.2%	$2.10Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
13	GPT-6 Astra	Low	31.1%	$2.44Cost per task	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
14	Claude Opus 5.5	Med	30.1%	$1.70Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
15	Claude Opus 5.5	Low	14.7%	$0.75Cost per task	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

Agentic codingFrontierSWE 2

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	69.9%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	GPT-6 Astra	not stated	65.5%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Opus 5.5	not stated	62.3%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

official results (Claude Sonnet 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	61.9%	—	—	Proximal	Claude Sonnet 5.5 System CardRetrieved 2026-11-14

official results (Claude Haiku 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Haiku 5.5	Max	43.8%	—	—	Proximal	Claude Haiku 5.5 System CardRetrieved 2026-11-14

Agentic codingCursorBench 4.0

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	57.8%	$13.43Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	High	56.0%	$3.97Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Xhigh	56.0%	$6.99Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
4	Claude Opus 5.5	Med	52.5%	$2.90Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
5	Claude Opus 5.5	Low	43.7%	$1.18Cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Introducing Claude Sonnet 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	57.8%	$13.43Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	High	56.0%	$3.97Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Xhigh	56.0%	$6.98Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
4	Claude Sonnet 5.5	Max	55.5%	$9.67Cost per task	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
5	Claude Sonnet 5.5	Xhigh	53.1%	$3.88Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
6	Claude Opus 5.5	Med	52.5%	$2.91Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
7	Claude Sonnet 5.5	High	47.8%	$1.67Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
8	Claude Opus 5.5	Low	43.7%	$1.17Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
9	Claude Sonnet 5.5	Med	39.2%	$0.70Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
10	Claude Sonnet 5.5	Low	35.8%	$0.50Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	60.0%	—	—	Cursor	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
2	Claude Fable 5.2	High	56.9%	—	—	Cursor	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
3	Claude Fable 5.2	Xhigh	56.3%	—	—	Cursor	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
4	Claude Fable 5.2	Med	54.8%	—	—	Cursor	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
5	Claude Fable 5.2	Low	53.2%	—	—	Cursor	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

CodingMirrorCode

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	97.8%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
2	Claude Opus 5.5	Max	95.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
3	Claude Haiku 5.5	Max	78.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
3	Claude Opus 5.5	High	78.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
5	GPT-6.1 Terra	Max	76.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
6	GPT-6.1 Sol	Max	61.1%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
7	GPT-6 Astra	Max	60.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
8	GPT-6 Astra	High	46.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12
9	GPT-6 Luna	Max	26.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0Retrieved 2026-11-12

CodingLMArena Text Coding (without style control)

third-party results (lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	High	1,542	—	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
2	Claude Sonnet 5.5	Xhigh	1,500	—	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
3	GPT-6 Astra	Max	1,494	—	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
4	GPT-6.1 Sol	Max	1,486	—	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
5	Claude Haiku 5.5	High	1,454	—	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
6	GPT-6 Luna	Max	1,432	—	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12

ReasoningHumanity's Last Exam (with tools)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	74.2%	$4.51Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Fable 5.2	Xhigh	73.4%	$1.48Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Fable 5.2	High	71.5%	$0.63Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
4	Claude Fable 5.2	Med	68.6%	$0.21Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
5	Claude Fable 5.2	Low	68.4%	$0.19Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
6	Claude Opus 5.5	Max	67.7%	$2.18Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
7	Claude Opus 5.5	Xhigh	66.4%	$0.69Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
8	Claude Opus 5.5	High	63.9%	$0.29Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
9	Claude Opus 5.5	Med	63.0%	$0.17Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
10	Claude Opus 5.5	Low	57.4%	$0.08Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
11	GPT-6 Astra	not stated	57.2%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	67.7%	—	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	GPT-6 Astra	not stated	57.2%	—	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Introducing Claude Sonnet 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	67.7%	—	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	not stated	64.5%	—	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	not stated	64.5%	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Haiku 5.5	not stated	57.4%	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12

ReasoningHumanity's Last Exam (without tools)

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	56.9%	$0.89Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	Xhigh	53.0%	$0.20Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	Claude Sonnet 5.5	High	47.9%	$0.08Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
4	Claude Haiku 5.5	Max	45.9%	$0.21Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
5	Claude Haiku 5.5	Xhigh	44.1%	$0.06Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
6	Claude Sonnet 5.5	Med	43.2%	$0.03Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
7	Claude Sonnet 5.5	Low	41.4%	$0.03Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
8	Claude Haiku 5.5	High	39.7%	$0.02Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
9	Claude Haiku 5.5	Med	35.5%	$0.007Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
10	Claude Haiku 5.5	Low	30.9%	$0.003Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	73.2%	—	—	Anthropic	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

official results (Claude Opus 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	64.4%	—	—	Anthropic	Claude Opus 5.5 System CardRetrieved 2026-11-14

ReasoningGPQA Diamond

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6 Astra	Max	95.8%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
2	Claude Sonnet 5.5	Max	95.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
3	GPT-6.1 Sol	Max	95.4%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
4	GPT-6.1 Terra	Max	94.3%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
5	Claude Fable 5.2	Max	93.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
6	Claude Opus 5.5	Max	90.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
7	GPT-6 Luna	Max	90.5%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12
8	Claude Haiku 5.5	Max	89.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0Retrieved 2026-11-12

ReasoningChess Puzzles(Epoch AI)

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	79.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
2	GPT-6 Astra	Max	74.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
3	GPT-6.1 Sol	Max	68.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
4	Claude Opus 5.5	Max	67.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
5	Claude Sonnet 5.5	Max	59.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
6	Claude Haiku 5.5	Max	55.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
7	GPT-6 Luna	Max	36.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12
8	Claude Haiku 5.5	Low	24.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0Retrieved 2026-11-12

MathFrontierMath Tiers 1-3(v2)

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	95.1%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
2	GPT-6.1 Sol	Max	93.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
2	GPT-6 Astra	Max	93.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
4	Claude Opus 5.5	Max	91.2%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
5	GPT-6.1 Terra	Max	90.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
6	Claude Sonnet 5.5	Max	88.8%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
7	GPT-6 Luna	Max	79.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12
8	Claude Haiku 5.5	Max	75.1%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0Retrieved 2026-11-12

MathFrontierMath Tier 4(v2)

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
1	GPT-6.1 Sol	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
3	GPT-6 Astra	High	97.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
3	GPT-6 Astra	Xhigh	97.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
3	GPT-6 Astra	Max	97.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
6	GPT-6 Astra	Med	97.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
7	Claude Opus 5.5	Max	95.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
8	GPT-6.1 Terra	Max	92.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
9	GPT-6 Astra	Low	87.8%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
10	GPT-6 Astra	not stated	82.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
11	Claude Sonnet 5.5	Max	80.5%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
12	GPT-6 Luna	Max	56.1%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12
13	Claude Haiku 5.5	Max	46.3%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0Retrieved 2026-11-12

MathOTIS Mock AIME 2024-2025

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6.1 Terra	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
1	GPT-6.1 Sol	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
1	Claude Sonnet 5.5	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
1	Claude Opus 5.5	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
1	GPT-6 Astra	Max	100.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
6	Claude Fable 5.2	Max	99.7%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
7	Claude Haiku 5.5	Xhigh	98.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
7	GPT-6 Luna	Max	98.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12
9	Claude Haiku 5.5	High	97.2%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0Retrieved 2026-11-12

KnowledgeGDPval-AA v2.1

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	1,846	$8.92Estimated cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Xhigh	1,820	$4.21Estimated cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
3	Claude Opus 5.5	High	1,692	$1.54Estimated cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
4	Claude Opus 5.5	Med	1,576	$0.86Estimated cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
5	GPT-6 Astra	Max	1,542	$4.53Estimated cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
6	GPT-6 Astra	Xhigh	1,516	$3.04Estimated cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
7	GPT-6 Astra	High	1,485	$2.43Estimated cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
8	GPT-6 Astra	Med	1,468	$1.82Estimated cost per task	—	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
9	GPT-6 Astra	Low	1,366	$0.85Estimated cost per task	frontier	not stated	Introducing Claude Opus 5.5Retrieved 2026-11-12
10	Claude Opus 5.5	Low	1,224	$0.21Estimated cost per task	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	1,840	$6.78Cost per task	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	Xhigh	1,731	$1.88Cost per task	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	Claude Haiku 5.5	Max	1,620	$0.87Cost per task	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
4	Claude Sonnet 5.5	High	1,551	$0.62Cost per task	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
5	Claude Haiku 5.5	Xhigh	1,513	$0.27Cost per task	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
6	GPT-6 Luna	Max	1,437	$0.09Cost per task	frontier	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
7	Claude Haiku 5.5	High	1,420	$0.09Cost per task	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
8	GPT-6 Luna	Xhigh	1,364	$0.05Cost per task	frontier	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
9	GPT-6 Luna	High	1,344	$0.03Cost per task	frontier	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
10	Claude Sonnet 5.5	Med	1,324	$0.27Cost per task	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
11	Claude Haiku 5.5	Med	1,277	$0.03Cost per task	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
12	GPT-6 Luna	Med	1,262	$0.02Cost per task	frontier	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
13	Claude Sonnet 5.5	Low	1,179	$0.22Cost per task	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
14	Claude Haiku 5.5	Low	1,125	$0.01Cost per task	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
15	GPT-6 Luna	Low	1,036	$0.004Cost per task	frontier	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12

official results (Introducing Claude Sonnet 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	1,846	—	—	Artificial Analysis	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	not stated	1,844	—	—	Artificial Analysis	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	1,831	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
2	Claude Fable 5.2	Xhigh	1,751	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
3	GPT-6.1 Sol	not stated	1,609	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

KnowledgeAA-Briefcase v1.1(Elo)

official results (Introducing Claude Sonnet 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	1,822	$21.05Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	Max	1,811	$29.19Cost per task	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
3	Claude Opus 5.5	Xhigh	1,780	$12.27Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
4	Claude Sonnet 5.5	Xhigh	1,746	$9.63Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
5	Claude Opus 5.5	High	1,705	$6.27Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
6	Claude Opus 5.5	Med	1,642	$4.40Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
7	Claude Sonnet 5.5	High	1,634	$3.95Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
8	Claude Sonnet 5.5	Med	1,461	$1.64Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
9	Claude Opus 5.5	Low	1,285	$1.15Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
10	Claude Sonnet 5.5	Low	1,264	$0.87Cost per task	frontier	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	not stated	1,824	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Haiku 5.5	not stated	1,578	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	GPT-6 Luna	not stated	1,336	—	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	1,800	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Fable 5.2	not stated	1,785	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	1,569	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6.1 Sol	not stated	1,555	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

KnowledgeAA-Briefcase v1.1(rubric)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	66.1%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Opus 5.5	not stated	61.2%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	52.0%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

KnowledgeRemote Labor Index(RLI)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	High	41.3%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Opus 5.5	not stated	21.3%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	20.8%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

KnowledgeHealthBench Professional (length-adjusted score)

official results (GPT-6 Astra System Card (Table 6. HealthBench Evaluations))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6 Astra	not stated	64.7	—	—	OpenAI	GPT-6 Astra System Card (Table 6. HealthBench Evaluations)Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	67.7	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
2	Claude Sonnet 5.5	Max	63	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
3	Claude Opus 5.5	Max	59.9	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14
4	Claude Haiku 5.5	Max	58.4	—	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

official results (Addendum to GPT-6 Astra System Card: GPT-6.1 Terra)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6.1 Sol	not stated	64.2	—	—	OpenAI	Addendum to GPT-6 Astra System Card: GPT-6.1 TerraRetrieved 2026-11-14
2	GPT-6.1 Terra	not stated	62.7	—	—	OpenAI	Addendum to GPT-6 Astra System Card: GPT-6.1 TerraRetrieved 2026-11-14
3	GPT-6 Luna	not stated	60.8	—	—	OpenAI	Addendum to GPT-6 Astra System Card: GPT-6.1 TerraRetrieved 2026-11-14

KnowledgeHealthBench Hard (length-adjusted score)

official results (GPT-6 Astra System Card (Table 6. HealthBench Evaluations))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6 Astra	not stated	36.6	—	—	OpenAI	GPT-6 Astra System Card (Table 6. HealthBench Evaluations)Retrieved 2026-11-12

official results (Addendum to GPT-6 Astra System Card: GPT-6.1 Terra)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6.1 Sol	not stated	36.2	—	—	OpenAI	Addendum to GPT-6 Astra System Card: GPT-6.1 TerraRetrieved 2026-11-14
2	GPT-6.1 Terra	not stated	31.4	—	—	OpenAI	Addendum to GPT-6 Astra System Card: GPT-6.1 TerraRetrieved 2026-11-14
2	GPT-6 Luna	not stated	31.4	—	—	OpenAI	Addendum to GPT-6 Astra System Card: GPT-6.1 TerraRetrieved 2026-11-14

KnowledgeSimpleQA Verified

third-party results (Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	94.0%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
2	GPT-6 Astra	Max	75.6%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
3	GPT-6.1 Sol	Max	73.9%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
4	Claude Opus 5.5	Max	72.2%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
5	GPT-6.1 Terra	Max	57.3%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
6	Claude Sonnet 5.5	Max	46.5%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
7	GPT-6 Luna	Max	41.4%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12
8	Claude Haiku 5.5	Max	23.8%	—	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0Retrieved 2026-11-12

MultimodalChartography (with tools)

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	89.0%	—	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Claude Fable 5.2 and Claude Mythos 5.2 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	91.6%	—	—	Anthropic	Claude Fable 5.2 and Claude Mythos 5.2 System CardRetrieved 2026-11-14

official results (Claude Sonnet 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	90.2%	—	—	Anthropic	Claude Sonnet 5.5 System CardRetrieved 2026-11-14

official results (Claude Haiku 5.5 System Card)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Haiku 5.5	Max	86.2%	—	—	Anthropic	Claude Haiku 5.5 System CardRetrieved 2026-11-14

MultimodalChartography (without tools)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	73.4%	$0.75Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Fable 5.2	Xhigh	71.4%	$0.15Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	71.0%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
4	Claude Fable 5.2	High	69.8%	$0.07Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
5	Claude Fable 5.2	Med	66.4%	$0.03Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
6	Claude Opus 5.5	Max	64.4%	$0.49Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
7	Claude Fable 5.2	Low	64.2%	$0.03Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
8	Claude Opus 5.5	Xhigh	62.4%	$0.08Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
9	Claude Opus 5.5	High	59.2%	$0.03Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
10	Claude Opus 5.5	Med	59.0%	$0.03Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
11	Claude Opus 5.5	Low	57.0%	$0.02Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

official results (Introducing Claude Sonnet 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	64.4%	—	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	not stated	61.6%	—	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	not stated	61.6%	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Haiku 5.5	not stated	46.4%	—	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	GPT-6 Luna	not stated	29.1%	—	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12

Tool useAutomationBench

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	GPT-6 Astra	Max	41.4%	$1.77Cost per task	frontier	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Max	40.0%	$1.37Cost per task	frontier	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
3	GPT-6 Astra	Xhigh	39.0%	$1.53Cost per task	—	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
4	GPT-6 Astra	High	37.1%	$1.45Cost per task	—	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
5	Claude Opus 5.5	Xhigh	34.4%	$0.86Cost per task	frontier	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
6	GPT-6 Astra	Med	34.1%	$1.28Cost per task	—	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
7	Claude Opus 5.5	High	32.0%	$0.70Cost per task	frontier	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
8	GPT-6 Astra	Low	30.3%	$1.08Cost per task	—	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
9	Claude Opus 5.5	Med	28.6%	$0.64Cost per task	frontier	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12
10	Claude Opus 5.5	Low	23.3%	$0.50Cost per task	frontier	Zapier	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	51.4%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Opus 5.5	not stated	42.5%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	41.4%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

Tool useWANDR

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	72.3%	$37.92Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
2	Claude Opus 5.5	Xhigh	71.3%	$29.06Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
3	Claude Opus 5.5	High	67.3%	$16.88Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
4	Claude Opus 5.5	Med	62.8%	$11.20Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12
5	Claude Opus 5.5	Low	31.2%	$1.20Cost per attempt	frontier	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12

Tool useWANDR(relaxed score)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	82.1%	$53.11Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Fable 5.2	Xhigh	80.6%	$39.84Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Fable 5.2	High	79.3%	$29.20Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
4	Claude Fable 5.2	Med	73.6%	$15.71Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
5	Claude Opus 5.5	Max	72.3%	$38.68Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
6	Claude Fable 5.2	Low	71.4%	$12.66Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
7	Claude Opus 5.5	Xhigh	71.3%	$29.25Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
8	Claude Opus 5.5	High	67.3%	$17.31Cost per attempt	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
9	Claude Opus 5.5	Med	62.8%	$11.40Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
10	Claude Opus 5.5	Low	31.2%	$1.25Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

Computer useOSWorld 2.1(partial)

official results (Introducing Claude Opus 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	81.8%	—	—	Anthropic	Introducing Claude Opus 5.5Retrieved 2026-11-12

official results (Introducing Claude Sonnet 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	81.8%	—	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	not stated	80.1%	—	—	Anthropic	Introducing Claude Sonnet 5.5Retrieved 2026-11-12

Computer useOSWorld 2.1(offline subset)

official results (Introducing Claude Haiku 5.5)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Sonnet 5.5	Max	83.9%	$5.73Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
2	Claude Sonnet 5.5	Xhigh	81.1%	$2.22Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
3	Claude Sonnet 5.5	High	73.2%	$1.38Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
4	Claude Haiku 5.5	Max	72.4%	$0.61Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
5	Claude Haiku 5.5	Xhigh	67.6%	$0.28Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
6	Claude Sonnet 5.5	Med	66.0%	$0.93Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
7	Claude Haiku 5.5	High	61.3%	$0.18Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
8	Claude Sonnet 5.5	Low	57.9%	$0.68Cost per attempt	—	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
9	Claude Haiku 5.5	Med	53.3%	$0.13Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
10	GPT-6 Luna	Max	48.9%	$0.20Cost per attempt	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
11	GPT-6 Luna	Xhigh	44.8%	$0.17Cost per attempt	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
12	GPT-6 Luna	High	42.3%	$0.14Cost per attempt	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
13	Claude Haiku 5.5	Low	42.0%	$0.07Cost per attempt	frontier	Anthropic	Introducing Claude Haiku 5.5Retrieved 2026-11-12
14	GPT-6 Luna	Med	37.5%	$0.13Cost per attempt	—	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12
15	GPT-6 Luna	Low	19.2%	$0.04Cost per attempt	frontier	not stated	Introducing Claude Haiku 5.5Retrieved 2026-11-12

Computer useOSWorld 2.1 (offline set, partial)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	89.3%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Opus 5.5	not stated	87.2%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	79.5%	—	—	not stated	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

WritingLMArena Text (style control, overall)

third-party results (lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	High	1,506	—	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
2	GPT-6.1 Sol	Max	1,482	—	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
3	GPT-6 Astra	Max	1,474	—	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
4	Claude Sonnet 5.5	Xhigh	1,473	—	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
5	GPT-6 Luna	Max	1,441	—	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12
6	Claude Haiku 5.5	High	1,422	—	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)Retrieved 2026-11-12

OtherBenchCAD 2.1 (with tools)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	Max	83.5	$50.77Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Fable 5.2	Xhigh	74.1	$23.23Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Opus 5.5	Max	65.3	$19.86Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
4	Claude Fable 5.2	High	61	$9.99Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
5	Claude Opus 5.5	Xhigh	54.5	$8.91Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
6	Claude Fable 5.2	Med	32.4	$2.66Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
7	Claude Opus 5.5	High	29.7	$2.57Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
8	Claude Fable 5.2	Low	26.9	$1.89Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
9	Claude Opus 5.5	Med	19.1	$1.11Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
10	Claude Opus 5.5	Low	6.3	$0.03Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

OtherCVE Rediscovery

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	Max	81.3%	$2.98Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Opus 5.5	Xhigh	76.4%	$1.05Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	Claude Opus 5.5	High	70.0%	$0.43Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
4	Claude Opus 5.5	Med	66.1%	$0.28Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
5	Claude Opus 5.5	Low	59.4%	$0.11Cost per attempt	frontier	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

OtherCyScenarioBench

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	67.6%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

OtherDe novo binder design (24 hours)

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Opus 5.5	not stated	82.6	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

OtherRoboDojo Sim

official results (Introducing Claude Fable 5.2 and Claude Mythos 5.2)

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	64.9%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
2	Claude Opus 5.5	not stated	39.4%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12
3	GPT-6 Astra	not stated	29.8%	—	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2Retrieved 2026-11-12

OtherEpoch Capabilities Index(ECI)

third-party results (Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0))

Rank	Model	effort	Score	Source cost (metric)	frontier	measured/reported by	source
1	Claude Fable 5.2	not stated	173.44	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
2	GPT-6 Astra	not stated	169	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
3	Claude Opus 5.5	not stated	168.54	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
4	GPT-6.1 Sol	not stated	168.48	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
5	GPT-6.1 Terra	not stated	167.67	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
6	Claude Sonnet 5.5	not stated	166.56	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
7	Claude Haiku 5.5	not stated	160.43	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12
8	GPT-6 Luna	not stated	156.41	—	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)Retrieved 2026-11-12