modelbench日本語

Claude Opus 4.5

AnthropicLegacy

Input price
$5.00
Per 1 million tokens (USD)
Output price
$25.00
Per 1 million tokens (USD)
Effort levels
3total
Default: High
Benchmark results
8total
Benchmark types: 8

Model information

API model ID
claude-opus-4-5-20251101
release date
Unknown
effort
Low / Med / High (default)
Effort source
Based on Anthropic effort documentation (platform.claude.com/docs/en/build-with-claude/effort).
Claude Code
unknown
Codex
unavailable
Availability notes
Opus 4.5 is absent from the Claude Code model settings page (checked 2026-11-12), and whether it can be selected could not be confirmed, so claudeCode is null. effort is included in supportedModels but absent from the xhigh/max support list, and is used with budget_tokens (extended thinking). The default effort is based on "Most Claude models default to high". Codex's official model list (learn.chatgpt.com/docs/models) contains only OpenAI models, so codex is false (connections via a custom provider are not counted as official options).
Availability source
https://code.claude.com/docs/en/model-config

Price (USD per 1 million tokens)

Prices do not vary by effort; token consumption does.

input$5.00
output$25.00
cache read$0.50
cache write$6.25

cacheWritePerMTok is for a 5-minute cache write. 1h write $10. 200K context.

Claude Opus 4.5 | $5 / MTok | $6.25 / MTok | $10 / MTok | $0.50 / MTok | $25 / MTok

Source: https://platform.claude.com/docs/en/about-claude/pricing

Benchmark results

BenchmarkeffortScoreSource cost (metric)measured/reported bysourceConditions and extraction method
GPQA Diamondnot statednot stated80.7%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Opus 4.5 (no thinking)". No effort suffix in Model version. Data retrieved 2026-11-12. No effort because effort is not stated in the source.transcribed from table
Source excerptgpqa_diamond.csv: Model version=claude-opus-4-5-20251101, Best score (across scorers)=0.8068181818181818, stderr=0.023909479653766753, Started at=2025-11-24T19:57:20.151ZLocation gpqa_diamond.csv, data row 205 (file line 295), id=mEkZaYpiurQ6TEMAALK3su, Model version=claude-opus-4-5-20251101
OTIS Mock AIME 2024-2025not statednot stated48.1%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Opus 4.5 (no thinking)". No effort suffix in Model version. Data retrieved 2026-11-12. No effort because effort is not stated in the source.transcribed from table
Source excerptotis_mock_aime_2024_2025.csv: Model version=claude-opus-4-5-20251101, Best score (across scorers)=0.48055555555555557, stderr=0.06405336755004033, Started at=2025-11-24T20:47:49.364ZLocation otis_mock_aime_2024_2025.csv, data row 211 (file line 307), id=nSb2focDRMyPJXErk3MV94, Model version=claude-opus-4-5-20251101
Chess Puzzles(Epoch AI)not statednot stated5.0%—Epoch AIEpoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0third-party · Retrieved 2026-11-12Run by Epoch AI (third-party measurement). Source label "Claude Opus 4.5 (no thinking)". No effort suffix in Model version. Data retrieved 2026-11-12. No effort because effort is not stated in the source.transcribed from table
Source excerptchess_puzzles.csv: Model version=claude-opus-4-5-20251101, Best score (across scorers)=0.05, stderr=0.021904291355759033, Started at=2026-10-13T13:42:24.000ZLocation chess_puzzles.csv, data row 103 (file line 157), id=4afZ2hjLiJnPy5pD7TfE3K, Model version=claude-opus-4-5-20251101
Epoch Capabilities Index(ECI)not statednot stated151.92—Epoch AIEpoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)third-party · Retrieved 2026-11-12A per-model metric that combines results across multiple benchmarks and multiple efforts (not per effort). Calculated by Epoch AI.transcribed from table
Source excerpteci_scores.csv: Model=Claude Opus 4.5, eci=151.92, eci_ci_low=150.32, eci_ci_high=153.58Location epoch_capabilities_index/eci_scores.csv, data row 61 (file line 62), Model=Claude Opus 4.5
τ²-bench airline (v1.0.1)High (default)High84.0%$0.40Average cost in USD to run one trajectory in this domainSierraτ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (sierra-research/tau2-bench, MIT)third-party · Retrieved 2026-11-13τ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (submission_type standard, tau2_bench_version 1.0.1). Submitter "Sierra", model_organization "Anthropic": Run by Sierra, which operates τ²-bench (third-party measurement). Verified on the leaderboard (trajectories available, prompts not modified, no questions omitted). The value is pass^1 (%). reasoning_effort "high". User simulator gpt-5.2. 4 trial runs (based on "_gpt-5.2_4trials" in the trajectories filename). The notes and evaluation_date (2026-05-05) refer to banking_knowledge (AllTools) ("Banking_knowledge domain only"); the evaluation date, seed, and simulator effort for this domain are not stated.transcribed from table
Source excerptairline.pass_1=84.0; airline.cost=0.39919Location In submissions/claude-opus-4-5_sierra_2026-02-26/submission.json: results.airline.pass_1 (line 22 of file), results.airline.cost (line 26). model_name "Claude Opus 4.5". trajectory_files.airline (line 14), methodology.user_simulator (line 55). methodology.notes (line 56) and evaluation_date (line 53) refer to banking_knowledge (AllTools)
τ²-bench banking_knowledge (v1.0.1)High (default)High24.7%—Sierraτ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (sierra-research/tau2-bench, MIT)third-party · Retrieved 2026-11-13τ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (submission_type standard, tau2_bench_version 1.0.1). Submitter "Sierra", model_organization "Anthropic": Run by Sierra, which operates τ²-bench (third-party measurement). Verified on the leaderboard (trajectories available, prompts not modified, no questions omitted). The value is pass^1 (%). reasoning_effort "high". User simulator gpt-5.2 (reasoning_effort low). Search configuration retrieval_config "alltools". Evaluation date 2026-05-05. Notes summary: rescored on 2026-07-15 following a scoring correction in v1.0.1; 4 trial runs; seed 300.transcribed from table
Source excerptbanking_knowledge.pass_1=24.74Location In submissions/claude-opus-4-5_sierra_2026-02-26/submission.json: results.banking_knowledge.pass_1 (line 43 of file). model_name "Claude Opus 4.5". methodology.notes (line 56)
τ²-bench retail (v1.0.1)High (default)High79.6%$0.39Average cost in USD to run one trajectory in this domainSierraτ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (sierra-research/tau2-bench, MIT)third-party · Retrieved 2026-11-13τ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (submission_type standard, tau2_bench_version 1.0.1). Submitter "Sierra", model_organization "Anthropic": Run by Sierra, which operates τ²-bench (third-party measurement). Verified on the leaderboard (trajectories available, prompts not modified, no questions omitted). The value is pass^1 (%). reasoning_effort "high". User simulator gpt-5.2. 4 trial runs (based on "_gpt-5.2_4trials" in the trajectories filename). The notes and evaluation_date (2026-05-05) refer to banking_knowledge (AllTools) ("Banking_knowledge domain only"); the evaluation date, seed, and simulator effort for this domain are not stated.transcribed from table
Source excerptretail.pass_1=79.61; retail.cost=0.38695Location In submissions/claude-opus-4-5_sierra_2026-02-26/submission.json: results.retail.pass_1 (line 29 of file), results.retail.cost (line 33). model_name "Claude Opus 4.5". trajectory_files.retail (line 15), methodology.user_simulator (line 55). methodology.notes (line 56) and evaluation_date (line 53) refer to banking_knowledge (AllTools)
τ²-bench telecom (v1.0.1)High (default)High92.3%$0.72Average cost in USD to run one trajectory in this domainSierraτ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (sierra-research/tau2-bench, MIT)third-party · Retrieved 2026-11-13τ²-bench leaderboard submission claude-opus-4-5_sierra_2026-02-26 (submission_type standard, tau2_bench_version 1.0.1). Submitter "Sierra", model_organization "Anthropic": Run by Sierra, which operates τ²-bench (third-party measurement). Verified on the leaderboard (trajectories available, prompts not modified, no questions omitted). The value is pass^1 (%). reasoning_effort "high". User simulator gpt-5.2. 4 trial runs (based on "_gpt-5.2_4trials" in the trajectories filename). The notes and evaluation_date (2026-05-05) refer to banking_knowledge (AllTools) ("Banking_knowledge domain only"); the evaluation date, seed, and simulator effort for this domain are not stated.transcribed from table
Source excerpttelecom.pass_1=92.32; telecom.cost=0.71992Location In submissions/claude-opus-4-5_sierra_2026-02-26/submission.json: results.telecom.pass_1 (line 36 of file), results.telecom.cost (line 40). model_name "Claude Opus 4.5". trajectory_files.telecom (line 16), methodology.user_simulator (line 55). methodology.notes (line 56) and evaluation_date (line 53) refer to banking_knowledge (AllTools)

Related rankings

References