Claude Opus 5.5

AnthropicCurrent

Input price: $4.00; Per 1 million tokens (USD)
Output price: $20.00; Per 1 million tokens (USD)
Effort levels: 5total; Default: Med
Benchmark results: 111total; Benchmark types: 35

Model information

API model ID: claude-opus-5-5
release date: 2026-09-22
effort: Low / Med (default) / High / Xhigh / Max
Effort source: Based on Anthropic effort documentation (platform.claude.com/docs/en/build-with-claude/effort).
Claude Code: available
Codex: unavailable
Availability notes: `opus` / `default` (default for Pro/Max/Team/Enterprise/API) resolves to Opus 5.5 (as of v2.1.280). Claude Code default effort is medium. 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	$4.00
output	$20.00
cache read	$0.20
cache write	$5.00

cacheWritePerMTok is for a 5-minute cache write. 1h write $8. cache read is 0.05x. fast mode $8/$40 (announcement page).

Claude Opus 5.5 | $4 / MTok | $5 / MTok | $8 / MTok | $0.20 / MTok | $20 / MTok

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

Benchmark results

Benchmark	effort	Score	Source cost (metric)	measured/reported by	source	Conditions and extraction method
Terminal-Bench 4.0Low	Low	38.5%	$1.29Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs Cost: "Opus 5.5 · Low: 38.5% $1.29"
Terminal-Bench 4.0Low	Low	38.5%	$1.29Cost per attempt	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs. cost: "Opus 5.5 · Low: 38.5% $1.29"
Terminal-Bench 4.0Med (default)	Med	57.6%	$2.94Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs Cost: "Opus 5.5 · Med: 57.6% $2.94"
Terminal-Bench 4.0Med (default)	Med	57.6%	$2.94Cost per attempt	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs. cost: "Opus 5.5 · Med: 57.6% $2.94"
Terminal-Bench 4.0High	High	64.2%	$3.88Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs Cost: "Opus 5.5 · High: 64.2% $3.88"
Terminal-Bench 4.0High	High	64.2%	$3.88Cost per attempt	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs. cost: "Opus 5.5 · High: 64.2% $3.88"
Terminal-Bench 4.0Xhigh	Xhigh	66.4%	$7.35Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs Cost: "Opus 5.5 · Xhigh: 66.4% $7.35"
Terminal-Bench 4.0Xhigh	Xhigh	66.4%	$7.35Cost per attempt	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs. cost: "Opus 5.5 · Xhigh: 66.4% $7.35"
Terminal-Bench 4.0Max	Max	64.8%	$11.24Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs Cost: "Opus 5.5 · Max: 64.8% $11.24"
Terminal-Bench 4.0Max	Max	64.8%	$11.24Cost per attempt	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "Terminal-Bench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Terminal-Bench 4.0 — Accuracy vs. cost: "Opus 5.5 · Max: 64.8% $11.24"
Terminal-Bench-Science 0.1not stated	not stated	58.7%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 58.7%Location Table, Terminal-Bench-Science 0.1 row: 58.7
Terminal-Bench-Science 0.1Max	Max	58.7%	—	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: Unless otherwise noted, all Claude Opus 5.5 results use adaptive thinking at max effort. production safeguards enabled. SE ±3.5–5pt. The GPT-6 Astra figure is a reported value from OpenAI.transcribed from table Source excerpt 58.7%Location Table, Terminal-Bench-Science 0.1 row: 58.7
FrontierCode v1.1(main set)Low	Low	47.3%	$0.40Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs Cost: "Opus 5.5 · Low: 47.3% $0.40"
FrontierCode v1.1(main set)Low	Low	47.3%	$0.40Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs. cost: "Opus 5.5 · Low: 47.3% $0.40"
FrontierCode v1.1(main set)Med (default)	Med	54.6%	$0.80Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs Cost: "Opus 5.5 · Med: 54.6% $0.80"
FrontierCode v1.1(main set)Med (default)	Med	54.6%	$0.80Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs. cost: "Opus 5.5 · Med: 54.6% $0.80"
FrontierCode v1.1(main set)High	High	54.0%	$1.09Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs Cost: "Opus 5.5 · High: 54.0% $1.09"
FrontierCode v1.1(main set)High	High	54.0%	$1.09Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs. cost: "Opus 5.5 · High: 54.0% $1.09"
FrontierCode v1.1(main set)Xhigh	Xhigh	51.4%	$2.25Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs Cost: "Opus 5.5 · Xhigh: 51.4% $2.25"
FrontierCode v1.1(main set)Xhigh	Xhigh	51.4%	$2.25Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs. cost: "Opus 5.5 · Xhigh: 51.4% $2.25"
FrontierCode v1.1(main set)Max	Max	54.4%	$6.19Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs Cost: "Opus 5.5 · Max: 54.4% $6.19"
FrontierCode v1.1(main set)Max	Max	54.4%	$6.19Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Evidence (same value as comparison group Accuracy vs Cost figure announced by Anthropic (Opus 5.5 and Sonnet 5.5 pages))Vendor-reported value. A point in the figure "FrontierCode v1.1, main set / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode v1.1, main set — Accuracy vs. cost: "Opus 5.5 · Max: 54.4% $6.19"
FrontierCode 2.0Low	Low	14.7%	$0.75Cost per task	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode 2.0 — Accuracy vs. cost: "Opus 5.5 · Low: 14.7% $0.75"
FrontierCode 2.0Med (default)	Med	30.1%	$1.70Cost per task	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode 2.0 — Accuracy vs. cost: "Opus 5.5 · Med: 30.1% $1.70"
FrontierCode 2.0High	High	34.5%	$2.42Cost per task	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode 2.0 — Accuracy vs. cost: "Opus 5.5 · High: 34.5% $2.42"
FrontierCode 2.0Xhigh	Xhigh	43.4%	$5.18Cost per task	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode 2.0 — Accuracy vs. cost: "Opus 5.5 · Xhigh: 43.4% $5.18"
FrontierCode 2.0Max	Max	45.5%	$12.82Cost per task	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "FrontierCode 2.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt FrontierCode 2.0 — Accuracy vs. cost: "Opus 5.5 · Max: 45.5% $12.82"
FrontierSWE 2not stated	not stated	62.3%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 62.3%Location Table, FrontierSWE v2 row: 62.3
CursorBench 4.0Low	Low	43.7%	$1.18Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs Cost: "Opus 5.5 · Low: 43.7% $1.18"
CursorBench 4.0Low	Low	43.7%	$1.17Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs. cost: "Opus 5.5 · Low: 43.7% $1.17"
CursorBench 4.0Med (default)	Med	52.5%	$2.90Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs Cost: "Opus 5.5 · Med: 52.5% $2.90"
CursorBench 4.0Med (default)	Med	52.5%	$2.91Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs. cost: "Opus 5.5 · Med: 52.5% $2.91"
CursorBench 4.0High	High	56.0%	$3.97Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs Cost: "Opus 5.5 · High: 56.0% $3.97"
CursorBench 4.0High	High	56.0%	$3.97Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs. cost: "Opus 5.5 · High: 56.0% $3.97"
CursorBench 4.0Xhigh	Xhigh	56.0%	$6.99Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs Cost: "Opus 5.5 · Xhigh: 56.0% $6.99"
CursorBench 4.0Xhigh	Xhigh	56.0%	$6.98Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs. cost: "Opus 5.5 · Xhigh: 56.0% $6.98"
CursorBench 4.0Max	Max	57.8%	$13.43Cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs Cost: "Opus 5.5 · Max: 57.8% $13.43"
CursorBench 4.0Max	Max	57.8%	$13.43Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CursorBench 4.0 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CursorBench 4.0 — Accuracy vs. cost: "Opus 5.5 · Max: 57.8% $13.43"
GDPval-AA v2.1not stated	not stated	1,846	—	Artificial Analysis	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: For details on how we run our evaluations, see the Sonnet 5.5 System Card (only efforts explicitly stated in the table); footnote 3: run by Artificial Analysis on a pre-release deployment (structured outputs bug, now fixed)transcribed from table Source excerpt 1846Location Table, GDPval-AA v2.1 row: 1846
GDPval-AA v2.1Low	Low	1,224	$0.21Estimated cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "GDPval-AA v2.1 / Elo vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt GDPval-AA v2.1 — Elo vs Cost: "Opus 5.5 · Low: 1224 $0.21"
GDPval-AA v2.1Med (default)	Med	1,576	$0.86Estimated cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "GDPval-AA v2.1 / Elo vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt GDPval-AA v2.1 — Elo vs Cost: "Opus 5.5 · Med: 1576 $0.86"
GDPval-AA v2.1High	High	1,692	$1.54Estimated cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "GDPval-AA v2.1 / Elo vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt GDPval-AA v2.1 — Elo vs Cost: "Opus 5.5 · High: 1692 $1.54"
GDPval-AA v2.1Xhigh	Xhigh	1,820	$4.21Estimated cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "GDPval-AA v2.1 / Elo vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt GDPval-AA v2.1 — Elo vs Cost: "Opus 5.5 · Xhigh: 1820 $4.21"
GDPval-AA v2.1Max	Max	1,846	$8.92Estimated cost per task	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "GDPval-AA v2.1 / Elo vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt GDPval-AA v2.1 — Elo vs Cost: "Opus 5.5 · Max: 1846 $8.92"
AA-Briefcase v1.1(Elo)not stated	not stated	1,800	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 1800Location Table, AA-Briefcase v1.1 (Elo) row: 1800
AA-Briefcase v1.1(Elo)Low	Low	1,285	$1.15Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AA-Briefcase v1.1 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt AA-Briefcase v1.1 — Accuracy vs. cost: "Opus 5.5 · Low: 1285.0 $1.15"
AA-Briefcase v1.1(Elo)Med (default)	Med	1,642	$4.40Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AA-Briefcase v1.1 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt AA-Briefcase v1.1 — Accuracy vs. cost: "Opus 5.5 · Med: 1642.0 $4.40"
AA-Briefcase v1.1(Elo)High	High	1,705	$6.27Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AA-Briefcase v1.1 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt AA-Briefcase v1.1 — Accuracy vs. cost: "Opus 5.5 · High: 1705.0 $6.27"
AA-Briefcase v1.1(Elo)Xhigh	Xhigh	1,780	$12.27Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AA-Briefcase v1.1 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt AA-Briefcase v1.1 — Accuracy vs. cost: "Opus 5.5 · Xhigh: 1780.0 $12.27"
AA-Briefcase v1.1(Elo)Max	Max	1,822	$21.05Cost per task	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AA-Briefcase v1.1 / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt AA-Briefcase v1.1 — Accuracy vs. cost: "Opus 5.5 · Max: 1822.0 $21.05"
AA-Briefcase v1.1(rubric)not stated	not stated	61.2%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 61.2%Location Table, AA-Briefcase v1.1 (rubric) row: 61.2
AutomationBenchnot stated	not stated	42.5%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 42.5%Location Table, AutomationBench row: 42.5
AutomationBenchLow	Low	23.3%	$0.50Cost per task	Zapier	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AutomationBench / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; measured by: Zapier (Opus 5.5 is from a Zapier early access evaluation; the others are from the Zapier public leaderboard)transcribed from text Source excerpt AutomationBench — Accuracy vs Cost: "Opus 5.5 · Low: 23.3% $0.50"
AutomationBenchMed (default)	Med	28.6%	$0.64Cost per task	Zapier	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AutomationBench / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; measured by: Zapier (Opus 5.5 is from a Zapier early access evaluation; the others are from the Zapier public leaderboard)transcribed from text Source excerpt AutomationBench — Accuracy vs Cost: "Opus 5.5 · Med: 28.6% $0.64"
AutomationBenchHigh	High	32.0%	$0.70Cost per task	Zapier	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AutomationBench / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; measured by: Zapier (Opus 5.5 is from a Zapier early access evaluation; the others are from the Zapier public leaderboard)transcribed from text Source excerpt AutomationBench — Accuracy vs Cost: "Opus 5.5 · High: 32.0% $0.70"
AutomationBenchXhigh	Xhigh	34.4%	$0.86Cost per task	Zapier	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AutomationBench / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; measured by: Zapier (Opus 5.5 is from a Zapier early access evaluation; the others are from the Zapier public leaderboard)transcribed from text Source excerpt AutomationBench — Accuracy vs Cost: "Opus 5.5 · Xhigh: 34.4% $0.86"
AutomationBenchMax	Max	40.0%	$1.37Cost per task	Zapier	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "AutomationBench / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; measured by: Zapier (Opus 5.5 is from a Zapier early access evaluation; the others are from the Zapier public leaderboard)transcribed from text Source excerpt AutomationBench — Accuracy vs Cost: "Opus 5.5 · Max: 40.0% $1.37"
Humanity's Last Exam (with tools)not stated	not stated	67.7%	—	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: For details on how we run our evaluations, see the Sonnet 5.5 System Card (only efforts explicitly stated in the table)transcribed from table Source excerpt 67.7%Location Table, Humanity's Last Exam (with tools) row: 67.7
Humanity's Last Exam (with tools)Low	Low	57.4%	$0.08Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Humanity’s Last Exam (with tools) — Accuracy vs. cost: "Opus 5.5 · Low: 57.4% $0.08"
Humanity's Last Exam (with tools)Med (default)	Med	63.0%	$0.17Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Humanity’s Last Exam (with tools) — Accuracy vs. cost: "Opus 5.5 · Med: 63.0% $0.17"
Humanity's Last Exam (with tools)High	High	63.9%	$0.29Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Humanity’s Last Exam (with tools) — Accuracy vs. cost: "Opus 5.5 · High: 63.9% $0.29"
Humanity's Last Exam (with tools)Xhigh	Xhigh	66.4%	$0.69Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Humanity’s Last Exam (with tools) — Accuracy vs. cost: "Opus 5.5 · Xhigh: 66.4% $0.69"
Humanity's Last Exam (with tools)Max	Max	67.7%	$2.18Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Humanity’s Last Exam (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Humanity’s Last Exam (with tools) — Accuracy vs. cost: "Opus 5.5 · Max: 67.7% $2.18"
Humanity's Last Exam (with tools)Max	Max	67.7%	—	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: Unless otherwise noted, all Claude Opus 5.5 results use adaptive thinking at max effort. production safeguards enabled.transcribed from table Source excerpt 67.7%Location Table, Humanity's Last Exam (with tools) row: 67.7
Humanity's Last Exam (without tools)Max	Max	64.4%	—	Anthropic	Claude Opus 5.5 System Cardofficial · Retrieved 2026-11-14	Vendor-reported value. Default settings for Table 8.1.A: adaptive thinking, max effort, default sampling, averaged over 5 trials. without tools. Cost not stated.transcribed from table Source excerpt 64.4Location Table 8.1.A “Evaluation summary”, Humanity’s Last Exam / No tools row, own model column
OSWorld 2.1(partial)not stated	not stated	81.8%	—	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: For details on how we run our evaluations, see the Sonnet 5.5 System Card (only efforts explicitly stated in the table)transcribed from table Source excerpt 81.8%Location Table, OSWorld 2.1 (partial) row: 81.8
OSWorld 2.1(partial)Max	Max	81.8%	—	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: Unless otherwise noted, all Claude Opus 5.5 results use adaptive thinking at max effort. production safeguards enabled.transcribed from table Source excerpt 81.8%Location Table, OSWorld 2.1 (partial) row: 81.8
OSWorld 2.1 (offline set, partial)not stated	not stated	87.2%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 87.2%Location Table, OSWorld 2.1 (offline set, partial) row: 87.2
Chartography (with tools)Max	Max	89.0%	—	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: Unless otherwise noted, all Claude Opus 5.5 results use adaptive thinking at max effort. production safeguards enabled.transcribed from table Source excerpt 89.0%Location Table, Chartography (with tools) row: 89.0
Chartography (without tools)not stated	not stated	64.4%	—	Anthropic	Introducing Claude Sonnet 5.5official · Retrieved 2026-11-12	Vendor-reported value. table note: For details on how we run our evaluations, see the Sonnet 5.5 System Card (only efforts explicitly stated in the table)transcribed from table Source excerpt 64.4%Location Table, Chartography (no tools) row: 64.4
Chartography (without tools)Low	Low	57.0%	$0.02Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Chartography (no tools) — Accuracy vs. cost: "Opus 5.5 · Low: 57.0% $0.02"
Chartography (without tools)Med (default)	Med	59.0%	$0.03Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Chartography (no tools) — Accuracy vs. cost: "Opus 5.5 · Med: 59.0% $0.03"
Chartography (without tools)High	High	59.2%	$0.03Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Chartography (no tools) — Accuracy vs. cost: "Opus 5.5 · High: 59.2% $0.03"
Chartography (without tools)Xhigh	Xhigh	62.4%	$0.08Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Chartography (no tools) — Accuracy vs. cost: "Opus 5.5 · Xhigh: 62.4% $0.08"
Chartography (without tools)Max	Max	64.4%	$0.49Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "Chartography (no tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt Chartography (no tools) — Accuracy vs. cost: "Opus 5.5 · Max: 64.4% $0.49"
WANDRLow	Low	31.2%	$1.20Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR — Accuracy vs Cost: "Opus 5.5 · Low: 31.2% $1.20"
WANDRMed (default)	Med	62.8%	$11.20Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR — Accuracy vs Cost: "Opus 5.5 · Med: 62.8% $11.20"
WANDRHigh	High	67.3%	$16.88Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR — Accuracy vs Cost: "Opus 5.5 · High: 67.3% $16.88"
WANDRXhigh	Xhigh	71.3%	$29.06Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR — Accuracy vs Cost: "Opus 5.5 · Xhigh: 71.3% $29.06"
WANDRMax	Max	72.3%	$37.92Cost per attempt	Anthropic	Introducing Claude Opus 5.5official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR / Accuracy vs Cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR — Accuracy vs Cost: "Opus 5.5 · Max: 72.3% $37.92"
WANDR(relaxed score)Low	Low	31.2%	$1.25Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR (relaxed score) — Accuracy vs. cost: "Opus 5.5 · Low: 31.2% $1.25"
WANDR(relaxed score)Med (default)	Med	62.8%	$11.40Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR (relaxed score) — Accuracy vs. cost: "Opus 5.5 · Med: 62.8% $11.40"
WANDR(relaxed score)High	High	67.3%	$17.31Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR (relaxed score) — Accuracy vs. cost: "Opus 5.5 · High: 67.3% $17.31"
WANDR(relaxed score)Xhigh	Xhigh	71.3%	$29.25Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR (relaxed score) — Accuracy vs. cost: "Opus 5.5 · Xhigh: 71.3% $29.25"
WANDR(relaxed score)Max	Max	72.3%	$38.68Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "WANDR (relaxed score) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic); WANDR: Claude was run with offline web search/fetch, programmatic tool calling, code execution, and a 980k token task budget (different from the Perplexity public configuration)transcribed from text Source excerpt WANDR (relaxed score) — Accuracy vs. cost: "Opus 5.5 · Max: 72.3% $38.68"
BenchCAD 2.1 (with tools)Low	Low	6.3	$0.03Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt BenchCAD 2.1 (with tools) — Accuracy vs. cost: "Opus 5.5 · Low: 6.3 $0.03"
BenchCAD 2.1 (with tools)Med (default)	Med	19.1	$1.11Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt BenchCAD 2.1 (with tools) — Accuracy vs. cost: "Opus 5.5 · Med: 19.1 $1.11"
BenchCAD 2.1 (with tools)High	High	29.7	$2.57Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt BenchCAD 2.1 (with tools) — Accuracy vs. cost: "Opus 5.5 · High: 29.7 $2.57"
BenchCAD 2.1 (with tools)Xhigh	Xhigh	54.5	$8.91Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt BenchCAD 2.1 (with tools) — Accuracy vs. cost: "Opus 5.5 · Xhigh: 54.5 $8.91"
BenchCAD 2.1 (with tools)Max	Max	65.3	$19.86Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "BenchCAD 2.1 (with tools) / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt BenchCAD 2.1 (with tools) — Accuracy vs. cost: "Opus 5.5 · Max: 65.3 $19.86"
CVE RediscoveryLow	Low	59.4%	$0.11Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CVE Rediscovery / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CVE Rediscovery — Accuracy vs. cost: "Opus 5.5 · Low: 59.4% $0.11"
CVE RediscoveryMed (default)	Med	66.1%	$0.28Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CVE Rediscovery / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CVE Rediscovery — Accuracy vs. cost: "Opus 5.5 · Med: 66.1% $0.28"
CVE RediscoveryHigh	High	70.0%	$0.43Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CVE Rediscovery / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CVE Rediscovery — Accuracy vs. cost: "Opus 5.5 · High: 70.0% $0.43"
CVE RediscoveryXhigh	Xhigh	76.4%	$1.05Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CVE Rediscovery / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CVE Rediscovery — Accuracy vs. cost: "Opus 5.5 · Xhigh: 76.4% $1.05"
CVE RediscoveryMax	Max	81.3%	$2.98Cost per attempt	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. A point in the figure "CVE Rediscovery / Accuracy vs. cost" on the Anthropic announcement page. values were taken from each point's aria-label (DOM); the page shows them only in the figure; value for its own model (announced by Anthropic)transcribed from text Source excerpt CVE Rediscovery — Accuracy vs. cost: "Opus 5.5 · Max: 81.3% $2.98"
CyScenarioBenchnot stated	not stated	67.6%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted. Opus 5.5 without safeguards.transcribed from table Source excerpt 67.6%Location Table, CyScenarioBench row: 67.6% without safeguards
De novo binder design (24 hours)not stated	not stated	82.6	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted. Opus 5.5 without safeguards.transcribed from table Source excerpt 82.6Location Table, Protein design row: 82.6 without safeguards
RoboDojo Simnot stated	not stated	39.4%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted. ‡ highest-available effort, run internally.transcribed from table Source excerpt 39.4%Location Table, RoboDojo Sim row: 39.4
Remote Labor Index(RLI)not stated	not stated	21.3%	—	Anthropic	Introducing Claude Fable 5.2 and Claude Mythos 5.2official · Retrieved 2026-11-12	Vendor-reported value. table note: Scores for Fable 5.2, Fable 5.1, and Opus 5.5 show results with safeguards on unless otherwise noted. All scores show each model’s highest-available effort unless otherwise noted.transcribed from table Source excerpt 21.3%Location Table, Remote Labor Index (RLI) row: 21.3
HealthBench Professional (length-adjusted score)Max	Max	59.9	—	Artificial Analysis	Claude Fable 5.2 and Claude Mythos 5.2 System Cardofficial · Retrieved 2026-11-14	Vendor-reported value. Independently run by Artificial Analysis at Anthropic’s request. max, length-adjusted score, GPT-5.4 low grader. fallback enabled. Not comparable with internal runs in earlier Claude system cards (Opus 4.8 grader). Cost not stated.transcribed from text Source excerpt 59.9%Location §8.16.2 “HealthBench Professional (Artificial Analysis)”, body text, the model’s length-adjusted score in parentheses
GPQA DiamondMax	Max	90.6%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, gpqa_diamond.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt gpqa_diamond.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=0.9059343434343434, stderr=0.019434397190192135, Started at=2026-09-22T18:27:39.000Z
FrontierMath Tiers 1-3(v2)Max	Max	91.2%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tiers_1_3_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt frontiermath_tiers_1_3_v2.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=0.9122807017543859, stderr=0.016786209858442095, Started at=2026-09-22T18:27:39.000Z
FrontierMath Tier 4(v2)Max	Max	95.0%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, frontiermath_tier_4_v2.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt frontiermath_tier_4_v2.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=0.95, stderr=0.041175470771058854, Started at=2026-09-22T18:27:39.000Z
OTIS Mock AIME 2024-2025Max	Max	100.0%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, otis_mock_aime_2024_2025.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt otis_mock_aime_2024_2025.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=1.0, stderr=0.0, Started at=2026-09-22T18:27:39.000Z
SimpleQA VerifiedMax	Max	72.2%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, simpleqa_verified.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt simpleqa_verified.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=0.722, stderr=0.014174516461485273, Started at=2026-09-22T18:27:39.000Z
Chess Puzzles(Epoch AI)Max	Max	67.0%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, chess_puzzles.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt chess_puzzles.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=0.67, stderr=0.04725815626252609, Started at=2026-10-13T13:01:31.000Z
MirrorCodeHigh	High	78.9%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (high)". effort is indicated by the suffix _high in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt mirrorcode.csv: Model version=claude-opus-5-5_high, Best score (across scorers)=0.7889, stderr=0.0788, Started at=2026-09-22T16:42:00.000Z
MirrorCodeMax	Max	95.6%	—	Epoch AI	Epoch AI, Capabilities & benchmarking (benchmark_data.zip, mirrorcode.csv) CC BY 4.0third-party · Retrieved 2026-11-12	Run by Epoch AI (third-party measurement). Source label "Claude Opus 5.5 (max)". effort is indicated by the suffix _max in Epoch's Model version. Data retrieved 2026-11-12.transcribed from table Source excerpt mirrorcode.csv: Model version=claude-opus-5-5_max, Best score (across scorers)=0.9555555555555556, stderr=0.025545201491160446, Started at=2026-10-30T18:25:28.000Z
Epoch Capabilities Index(ECI)not stated	not stated	168.54	—	Epoch AI	Epoch AI, epoch_capabilities_index/eci_scores.csv (CC BY 4.0)third-party · Retrieved 2026-11-12	A per-model metric that combines results across multiple benchmarks and multiple efforts (not per effort). Calculated by Epoch AI.transcribed from table Source excerpt eci_scores.csv: Model=Claude Opus 5.5, eci=168.54, eci_ci_low=165.63, eci_ci_high=171.83
LMArena Text (style control, overall)High	High	1,506	—	LMArena	lmarena-ai/leaderboard-dataset text_style_control/latest-00000-of-00001.parquet (CC BY 4.0)third-party · Retrieved 2026-11-12	Human head-to-head votes (Bradley-Terry). Source label "Claude Opus 5.5 (High)" (effort is in parentheses in the label). Release date 2026-11-09.transcribed from table Source excerpt model_name=Claude Opus 5.5 (High), rating=1506.07, rating_lower=1500.0, rating_upper=1512.1, vote_count=15249, rank=4, category=overall, leaderboard_publish_date=2026-11-09
LMArena Text Coding (without style control)High	High	1,542	—	LMArena	lmarena-ai/leaderboard-dataset text/latest-00000-of-00001.parquet (CC BY 4.0)third-party · Retrieved 2026-11-12	Human head-to-head votes (Bradley-Terry). Source label "Claude Opus 5.5 (High)" (effort is in parentheses in the label). Release date 2026-11-09.transcribed from table Source excerpt model_name=Claude Opus 5.5 (High), rating=1542.21, rating_lower=1531.1, rating_upper=1553.3, vote_count=3339, rank=2, category=coding, leaderboard_publish_date=2026-11-09

Claude Opus 5.5

Model information

Price (USD per 1 million tokens)

Benchmark results

Related rankings

References