whatmodel.sh / Claude Opus 4.7

Claude Opus 4.7

+ contribute

Claude Opus 4.7 is a closed-weights language model from Anthropic, released April 2026.

75.0±0.5#40 of 719 LLMsfrom 42 benchmark results across 15 sources

$ how it compares

The LLMs ranked either side of Claude Opus 4.7, out of 828 with a score, then more from Anthropic.

#	model	score	where	released
39	Gemini 3.1 Pro (Preview) Google	75±1	api $4.50/M	2026-02
40	Muse Spark 1.2 Meta	75±1	api $2.00/M	2026-08
41	Claude Opus 4.7 Anthropic	75±1	api $10.00/M	2026-04
42	Grok 4.5 SpaceXAI (xAI)	75±1	api $3.00/M	2026-07
43	Muse Spark 1.1 Meta	75±1	api $2.00/M	2026-07
	more from Anthropic
2	Claude Opus 5.5 Anthropic	87±1	api $8.00/M	2026-09
4	Claude Fable 5.1 Anthropic	85±1	api $20.00/M	2026-09
5	Claude Sonnet 5.5 Anthropic	84±1	api $4.00/M	2026-09

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #34 of 300 · 13% of its score · 7 tests

Beats GPT-6 Luna, GPT-5.6 Luna, Gemini 3.7 Flash here, though they rank higher overall. Trails GLM-5.3-Flash, Grok 4.7 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Verified	83.5% ✎	▁··▁·▁▂▁▄█▄▃ beats 29/29	hard · avg 19%	▇▇ 4.3%	sharp
Agents' Last Exam overall full-pass rate (fixed OpenClaw + GUI-as-Tool)	15.1% ✎	▄···█▄▄▄█▄·█ beats 8/10	very hard · avg 1.7%	▇ 2.5%	typical
τ³-bench banking_knowledge (v1.0.1 grading)	40.2% ✎	▃▃▂▃·▆▃▂█▅▅▂ beats 19/25	very hard · avg 4.1%	▇ 2.0%	at its level
Toolathlon (Tool Decathlon) lab	59.3% ✎	█···█··█·███ beats 3/5	hard · avg 16%	▇ 1.5%	at its level
SWE Atlas (Codebase Q&A, Test Writing, Refactoring; mini-SWE-agent) pass@1	38.9% ✎	▄·▄··█···▄·█ beats 6/6	very hard · avg 6.1%	▇ 1.4%	at its level
RWSearch	79.3% ✎	████··█····█ beats 3/5	medium · avg 44%	▇ 1.0%	typical
TerminalWorld-Verified (200 tasks) with Terminus-2	62.5% ✎	▄··▄·▄·█···▄ beats 5/5	hard · avg 27%	▇ 0.6%	blunt · at its level

coding▁▁▁▃▆██▇▆▄▃▁ #35 of 458 · 11% of its score · 3 tests

Beats GPT-5.5, GPT-5.6 Terra, GPT-5.4 here, though they rank higher overall. Trails Qwen3.8 27B, GLM-5.2, GLM-5.3-Flash here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Arena WebDev (Code Arena)	1557.3 ✎	▁▂▄▃▅█▄▇▆▄▁▁ beats 87/119	avg 981	▇▇ 4.7%	sharp
SWE-bench Verified	83.5% ✎	▁··▁·▁▂▁▄█▄▃ beats 29/29	hard · avg 19%	▇▇ 4.3%	sharp
SWE Atlas (Codebase Q&A, Test Writing, Refactoring; mini-SWE-agent) pass@1	38.9% ✎	▄·▄··█···▄·█ beats 6/6	very hard · avg 6.1%	▇ 1.4%	at its level

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #6 of 393 * · 3.0% of its score · 2 tests

Beats GPT-6 Astra, GPT-6.1 Sol, Claude Sonnet 5.5 here, though they rank higher overall. Trails Claude Opus 4.6 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
LMArena (text)	1501.3 ✎	▁▁▁▂▃▃▃▆▆▇█▃ beats 375/379	avg 1268	▇ 2.7%	blunt
IFEval (aggregation unspecified)	88.7% ✎	█·██·█···█·█ beats 1/5	easy · avg 74%	▇ 0.3%	maxed out

math▁▂▃▅▆▅▄█▄▁▁▁ #38 of 398 · 47% of its score · 14 tests

Trails GPT-5.2, Qwen3.7 Max, Claude Opus 4.6 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
FrontierMath tiers 1–3 v2	70.2% ✎	▂··▁▁▃▄▄█▅▃▃ beats 59/82	very hard · avg 1.2%	▇▇▇▇▇ 12%	sharp · at its level
FrontierMath tier 4 v2	31.7% ✎	▁▂▁▂█▅▂▂▁▁▁▁ beats 37/63	very hard · avg 0.1%	▇▇▇▇ 11%	sharp · at its level
MathArena ArXivMath April 2026 (40 problems)	57.5% ✎	▁·····▅██▄·▁ beats 12/20	very hard · avg 7.2%	▇▇ 6.0%	sharp · at its level
MathArena ArXivMath February 2026 (32 problems)	40.6% ✎	▁▁▁▁█▇▄▃▁▃·▃ beats 12/24	very hard · avg 9.1%	▇ 3.5%	at its level
MathArena ArXivMath March 2026 (30 problems)	50.8% ✎	█▃▅·█▃▃·▅·▃▃ beats 4/14	very hard · avg 15%	▇ 2.9%	at its level
MathArena BrokenArXiv April 2026 (61 problems)	4.1% ✎	▃▂▆█▃▃▂▂▂··▂ beats 0/18	very hard · avg 0.7%	▇ 2.4%	sharp · at its level · noisy
MathArena ArXivMath January 2026 (23 problems)	52.2% ✎	▂··▄▂▄█▆▆▄▄█ beats 6/23	hard · avg 22%	▇ 2.2%	at its level
MathArena BrokenArXiv March 2026 (56 problems)	5.8% ✎	█·▆█▂·▂▂···▂ beats 0/14	very hard · avg 0.8%	▇ 1.4%	sharp · noisy

6 more tests

test	result	vs the others in its run	difficulty	counts	why
MathArena BrokenArXiv February 2026 (31 problems)	4.0% ✎	▃▃▂▃█▃▂·▃··▂ beats 2/16	very hard · avg 0.7%	▇ 1.4%	sharp · noisy
MathArena ArXivLean March 2026 (41 problems)	14.6% ✎	▂····▂·····█ beats 3/5	very hard · avg 1.0%	▇ 1.0%	typical
IMO-AnswerBench	81.8% ✎	▁▁·▁▁▁▄▅█▄▂▃ beats 25/42	hard · avg 25%	▇ 1.0%	noisy
USAMO 2026 lab	69.3% ✎	█·██······██ beats 1/4	hard · avg 16%	▇ 0.9%	sharp · noisy
MathConstraint full release, no tools (329 instances)	42.6% ✎	▄··▄▄·▄▄█▄·▄ beats 5/8	very hard · avg 11%	▇ 0.7%	at its level
OTIS mock AIME 2024–2025	86.7% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 132/193	very hard · avg 8.5%	▇ 0.5%	sharp · maxed out · noisy

vision & documents▁▁▃▂▁▃▄▂▆▇█▄ #4 of 147 · 4.4% of its score · 3 tests

Beats GPT-6 Astra, GPT-6.1 Sol, Claude Fable 5.1 here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Arena vision (style controlled)	1297.7 ✎	▁▁▂▂▂▂▄▅▅▆██ beats 142/145	avg 1068	▇ 1.8%	blunt
CharXiv (reasoning questions) lab	81.3% ✎	███··█·██··█ beats 5/6	hard · avg 29%	▇ 1.7%	typical
Arena document (style controlled)	1493.0 ✎	▁▁▃▃▄▃▇█▅▇▄▅ beats 35/37	avg 1335	▇ 0.9%	blunt

web search▂▁▂▆▄█▃▃▁▁·▁ #30 of 80 * · 2.1% of its score · 2 tests

Beats Gemini 3.1 Pro (Preview) here, though they rank higher overall. Trails GPT-4o mini (2024-07-18), Mistral Small 3.1 24B (2503), GPT-4o (2024-08-06) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Arena search systems (style controlled)	1212.0 ✎	▁·▁▁▂▂▂█▅▂▁▁ beats 26/30	avg 1055	▇ 1.1%	blunt
RWSearch	79.3% ✎	████··█····█ beats 3/5	medium · avg 44%	▇ 1.0%	typical

other tests14 not in a task · 26% of its score

test	result	vs the others in its run	difficulty	counts	why
EBR-bench (Epoch)	19.0% ✎	▄▂·▆█▃▆▄▆▆·▄ beats 10/23	very hard · avg 0.3%	▇▇ 4.9%	sharp
Furniture Assembly (Epoch)	33.3% ✎	█▆▆▆▆▂▅▂▂▂·▅ beats 10/30	very hard · avg 1.2%	▇▇ 4.4%	sharp · at its level
Mystery Game Puzzles (Epoch)	28.0% ✎	█··▂▃▇▄█▂▃▁▁ beats 45/74	very hard · avg 9.4%	▇ 3.4%	sharp · noisy
Chess Puzzles (Epoch)	7.0% ✎	█▁▃▁▁▂▁▄▂▂▁▁ beats 52/135	very hard · avg 5.9%	▇ 2.2%	sharp · at its level · noisy
GPQA Diamond	86.4% ✎	▁▁▁▁▂▄▆▄▅▇█▅ beats 153/208	hard · avg 40%	▇ 2.1%	sharp
EQ-Bench 3 (emotional intelligence, judge Elo)	1460.9 ✎	▁·▂▁▄▅▅▃█▄▁▃ beats 62/62	avg 1044	▇ 1.6%	typical
WildClawBench	62.2% ✎	▃·▅▅█·▅█▅▃▃▃ beats 17/17	very hard · avg 11%	▇ 1.6%	at its level
The Token Games Elo (10-model puzzle-duel tournament)	1100.0 ✎	█··▄█··▄·▄·▄ beats 5/7	avg 815	▇ 1.5%	typical

6 more tests

test	result	vs the others in its run	difficulty	counts	why
FORTE (AGI-Eval)	77.6% ✎	▄···█·····▄█ beats 4/5	medium · avg 37%	▇ 1.1%	at its level
SimpleQA Verified	51.7% ✎	▁▁▃▂▃▃▆▆█▃▂▅ beats 56/78	very hard · avg 8.0%	▇ 1.1%	at its level
LongSocialBench (full-context, 4-option MCQ over discussion threads)	56.0% ✎	▆▂·▄█·▂▂·▄▄▃ beats 16/16	very hard · avg 36%	▇ 1.0%	blunt · at its level
WritingBench	85.3% ✎	▁·▁▂▁··▅█▂▂▁ beats 15/20	medium · avg 55%	▇ 1.0%	typical
MirrorCode (Epoch)	31.1% ✎	█·▄▄·▄·▄·▄·█ beats 4/8	very hard · avg 1.5%	▇ 0.5%	sharp · noisy
Vectara hallucination leaderboard (HHEM-2.3, factual consistency rate)	88.0% ✎	▁▂▁▂▅█▂▃▄▃▁▁ beats 24/98	very easy · avg 90%	▇ 0.0%	blunt

$ where to run it

Through an API

api · $10.00/M · $5 in / $25 out per 1M tokens · 1M context · docs.anthropic.com ↗ · price sourceStandard base token rate from models.dev. Provider-specific tiers, long-context surcharges, caching and promotions may differ.

$ how the score was worked out · Epoch AI — internal evaluatio… 46%, MathArena ArXivMath / BrokenA… 21%, LMArena leaderboard dataset 11% and 12 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

source	results	share of the score	licence
Epoch AI — internal evaluations	11	▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 46%	CC BY 4.0
MathArena ArXivMath / BrokenArXiv / ArXivLean model-output datasets	8	▇▇▇▇▇▇ 21%	CC BY-SA 4.0
LMArena leaderboard dataset	5	▇▇▇ 11%	CC BY 4.0
meituan-longcat/LongCat-2.0 model card	5	▇ 4%	mit
Anthropic system cards and model announcements (lab-reported scores) · lab-reported	3	▇ 4%	Factual vendor-reported score (not licensed content)
Agents' Last Exam — authors' own fixed-harness model comparison	1	▇ 3%	CC BY 4.0
τ²/τ³-bench leaderboard: Sierra's own evaluation runs	1	▇ 2%	MIT
EQ-Bench 3 canonical leaderboard results	1	▇ 2%	MIT
WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation (arXiv:2605.10912v1)	1	▇ 2%	CC BY 4.0
The Token Games: Evaluating Language Model Reasoning with Puzzle Duels — authors' own tournament	1	▇ 2%	CC BY-SA 4.0
SWE Atlas: Benchmarking Coding Agents Beyond Issue Resolution (arXiv:2605.08366v1)	1	▇ 1%	CC BY 4.0
LongSocialBench: Do Long-Context LLMs Understand Online Discussion Threads? (arXiv:2610.04118v1)	1	▇ 1%	CC BY 4.0
MathConstraint: Automated Generation of Verified Combinatorial Reasoning Instances for LLMs — authors' own model runs	1	▇ 1%	CC BY 4.0
TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks — authors' own Terminus-2 model comparison	1	▇ 1%	CC BY 4.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)	1	▇ 0%	Apache-2.0
prior · pull toward the middle		▇ 0%
weighted average		= 75.0