whatmodel.sh / Claude Opus 4.6

Claude Opus 4.6

+ contribute

Claude Opus 4.6 is a closed-weights language model from Anthropic, released February 2026.

74.7±0.5#45 of 719 LLMsfrom 53 benchmark results across 27 sources

$ how it compares

The LLMs ranked either side of Claude Opus 4.6, out of 828 with a score, then more from Anthropic.

#	model	score	where	RAM	released
46	Claude Sonnet 5 Anthropic	75±1	api $4.00/M		2026-06
47	Grok 4.7 SpaceXAI (xAI)	75±1	api $3.00/M		2026-09
48	Claude Opus 4.6 Anthropic	75±1	api $10.00/M		2026-02
49	GPT-5.3-Codex OpenAI	74±1	api $4.81/M		2026-02
50	DeepSeek V4 Flash (0731) DeepSeek	74±1	api $0.11/M +local	~92GB	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▁▁▃▆█▆▅▅▅▄▃▁ #47 of 300 · 20% of its score · 14 tests

Beats Gemini 3.1 Pro (Preview), Muse Spark 1.2, Muse Spark 1.1 here, though they rank higher overall. Trails Claude Sonnet 4.6, Qwen3.8 27B, GLM-5.2 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Verified	78.7% ✎	▁··▁·▁▂▁▄█▄▁ beats 27/29	hard · avg 19%	▇▇ 4.0%	sharp
Agents' Last Exam overall full-pass rate (fixed OpenClaw + GUI-as-Tool)	14.1% ✎	▄···█▄▄▄█▄·█ beats 6/10	very hard · avg 1.7%	▇ 2.2%	typical
τ³-bench banking_knowledge (v1.0.1 grading)	27.3% ✎	▃▃▂▃·▆▃▂█▅▅▂ beats 11/25	very hard · avg 4.1%	▇ 1.8%	at its level
E2E-SWE (build a working codebase from scratch, 186 tasks) pass@1	22.6% ✎	▃··▅··▃█▃▅▃▅ beats 2/12	very hard · avg 3.1%	▇ 1.7%	sharp · at its level
ContractBench success rate (33 observation-contract tool tasks)	77.8% ✎	▆··▆▄▂▄▂▆▆█▄ beats 23/23	very hard · avg 14%	▇ 1.7%	sharp · at its level
Vibe Code Bench (test split, 50 web apps)	53.5% ✎	▃··▃·▃██▃▃▃█ beats 12/14	very hard · avg 2.3%	▇ 1.4%	sharp · at its level · noisy
Toolathlon (Tool Decathlon) lab	56.8% ✎	█···█··█·███ beats 2/5	hard · avg 16%	▇ 1.4%	at its level
Terminal-Bench 2.0 lab	65.4% ✎	▂▂▅▅████▆▂▃▃ beats 30/36	very hard · avg 12%	▇ 1.3%	at its level · small run

6 more tests

test	result	vs the others in its run	difficulty	counts	why
SWE Atlas (Codebase Q&A, Test Writing, Refactoring; mini-SWE-agent) pass@1	31.8% ✎	▄·▄··█···▄·█ beats 4/6	very hard · avg 6.1%	▇ 1.3%	at its level
Thinkingbox-bench pass@1 (507 stateful business-workflow agent tasks, N=20)	32.1% ✎	▅··▃·▃·█▃▅██ beats 4/15	very hard · avg 4.9%	▇ 1.1%	at its level
RWSearch	81.3% ✎	████··█····█ beats 4/5	medium · avg 44%	▇ 0.9%	typical
EEGAgentBench overall (tool-based EEG analysis agents, 6 tasks)	46.3% ✎	▂·▁▁··▇▇█▂·▁ beats 13/25	hard · avg 29%	▇ 0.7%	blunt · at its level
τ²-bench Telecom lab	99.3% ✎	▁·▅█▆▅▂▃▃▂▃▃ beats 87/89	hard · avg 29%	▇ 0.7%	sharp · maxed out · noisy
T1-Bench multi-domain conversational agent pass^3	34.5% ✎	▂▂··▂·▂·▂▂█▄ beats 8/11	hard · avg 25%	▇ 0.0%	blunt · at its level · noisy

coding▁▁▁▃▆██▇▆▄▃▁ #37 of 458 · 13% of its score · 5 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)	1545.9 ✎	▁▂▄▃▅█▄▇▆▄▁▁ beats 84/119	avg 981	▇▇ 4.2%	sharp
SWE-bench Verified	78.7% ✎	▁··▁·▁▂▁▄█▄▁ beats 27/29	hard · avg 19%	▇▇ 4.0%	sharp
E2E-SWE (build a working codebase from scratch, 186 tasks) pass@1	22.6% ✎	▃··▅··▃█▃▅▃▅ beats 2/12	very hard · avg 3.1%	▇ 1.7%	sharp · at its level
Vibe Code Bench (test split, 50 web apps)	53.5% ✎	▃··▃·▃██▃▃▃█ beats 12/14	very hard · avg 2.3%	▇ 1.4%	sharp · at its level · noisy
SWE Atlas (Codebase Q&A, Test Writing, Refactoring; mini-SWE-agent) pass@1	31.8% ✎	▄·▄··█···▄·█ beats 4/6	very hard · avg 6.1%	▇ 1.3%	at its level

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #3 of 393 * · 2.7% of its score · 2 tests

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

test	result	vs the others in its run	difficulty	counts	why
LMArena (text)	1504.7 ✎	▁▁▁▂▃▃▃▆▆▇█▃ beats 378/379	avg 1268	▇ 2.4%	blunt
IFEval (aggregation unspecified)	92.2% ✎	█·██·█···█·█ beats 3/5	easy · avg 74%	▇ 0.3%	maxed out

math▁▂▃▅▆▅▄█▄▁▁▁ #37 of 398 · 33% of its score · 10 tests

Beats Claude Opus 4.7, Grok 4.7 here, though they rank higher overall. Trails GPT-5.2, Qwen3.7 Max here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
FrontierMath tiers 1–3 v2	66.0% ✎	▂··▁▁▃▄▄█▅▃▃ beats 53/82	very hard · avg 1.2%	▇▇▇▇▇ 11%	sharp · at its level
FrontierMath tier 4 v2	26.8% ✎	▁▂▁▂█▅▂▂▁▁▁▁ beats 30/63	very hard · avg 0.1%	▇▇▇▇ 9.2%	sharp · at its level
MathArena ArXivMath February 2026 (32 problems)	40.6% ✎	▁▁▁▁█▇▄▃▁▃·▃ beats 13/24	very hard · avg 9.1%	▇ 3.1%	at its level
MathArena ArXivMath March 2026 (30 problems)	62.5% ✎	█▃▅·█▃▃·▅·▃▃ beats 10/14	very hard · avg 15%	▇ 2.6%	at its level
MathArena ArXivMath January 2026 (23 problems)	72.8% ✎	▂··▄▂▄█▆▆▄▄█ beats 20/23	hard · avg 22%	▇ 2.0%	at its level
MathArena ArXivMath December 2025 (17 problems)	57.4% ✎	▃·▃▃█▁·▄·▃▁▁ beats 16/18	hard · avg 19%	▇ 1.5%	at its level
MathArena BrokenArXiv March 2026 (56 problems)	5.8% ✎	█·▆█▂·▂▂···▂ beats 1/14	very hard · avg 0.8%	▇ 1.2%	sharp · noisy
MathArena BrokenArXiv February 2026 (31 problems)	3.2% ✎	▃▂▂▃█▃▂·▃··▂ beats 1/16	very hard · avg 0.7%	▇ 1.2%	sharp · noisy

2 more tests

test	result	vs the others in its run	difficulty	counts	why
MathNet-Solve-Test micro-average accuracy (6,400 olympiad problems)	45.7% ✎	▃▃···█▃·▃·▃▃ beats 5/8	hard · avg 15%	▇ 0.8%	at its level
OTIS mock AIME 2024–2025	91.1% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 146/193	very hard · avg 8.5%	▇ 0.5%	sharp · maxed out · noisy

vision & documents▁▁▃▂▁▃▄▂▆▇█▄ #18 of 147 · 5.1% of its score · 5 tests

Beats Claude Sonnet 5.5, GPT-5.5, MiMo-V2.6-Pro here, though they rank higher overall. Trails Gemini 3.5 Flash here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Arena vision (style controlled)	1299.4 ✎	▁▁▂▂▂▂▄▅▅▆██ beats 143/145	avg 1068	▇ 1.6%	blunt
CharXiv (reasoning questions) lab	69.1% ✎	███··█·██··█ beats 1/6	hard · avg 29%	▇ 1.6%	typical
Arena document (style controlled)	1495.3 ✎	▁▁▃▃▄▃▇█▅▇▄▅ beats 36/37	avg 1335	▇ 0.8%	blunt
BEAR-Bench (bilingual RU/EN enterprise and academic document images, overall accuracy)	61.5% ✎	▃▃·▅·▃·█·█▅▅ beats 9/14	hard · avg 24%	▇ 0.8%	at its level
Diagram-MMU diagram question answering (DQA accuracy, scientific TikZ diagrams)	68.8% ✎	▄·····▄▄·█·█ beats 0/6	medium · avg 55%	▇ 0.3%	blunt

web search▂▁▂▆▄█▃▃▁▁·▁ #23 of 80 * · 1.9% of its score · 2 tests

Beats Claude Opus 4.8, GPT-5.4, Claude Haiku 5.5 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)	1223.3 ✎	▁·▁▁▂▂▂█▅▃▁▁ beats 27/30	avg 1055	▇ 1.0%	blunt
RWSearch	81.3% ✎	████··█····█ beats 4/5	medium · avg 44%	▇ 0.9%	typical

other tests19 not in a task · 33% of its score

test	result	vs the others in its run	difficulty	counts	why
EBR-bench (Epoch)	12.7% ✎	▄▂·▆█▂▆▄▆▆·▄ beats 7/23	very hard · avg 0.3%	▇▇ 4.3%	sharp
Furniture Assembly (Epoch)	28.3% ✎	█▆▆▆▆▂▅▂▂▂·▅ beats 8/30	very hard · avg 1.2%	▇▇ 3.8%	sharp · at its level
GPQA Diamond +1	88.4% ✎	▁▁▁▁▂▄▆▄▅▇█▅ beats 163/208	hard · avg 40%	▇▇ 3.8%	sharp · several runs
ARC-AGI-2 public evaluation set (ARC Prize model_baseline, pass@2)	79.0% ✎	█··▂▅▁▃▁▃·▁▃ beats 24/24	very hard · avg 0.3%	▇ 3.0%	sharp · at its level · noisy
Mystery Game Puzzles (Epoch)	25.0% ✎	█··▂▃▇▄█▂▃▁▁ beats 42/74	very hard · avg 9.4%	▇ 2.9%	sharp · noisy
Chess Puzzles (Epoch)	14.0% ✎	█▁▁▁▁▃▁▄▂▂▁▁ beats 66/135	very hard · avg 5.9%	▇ 1.9%	sharp · at its level · noisy
KINA (Knowledge Index of Noah's Ark) accuracy, 899 ten-option items	49.9% ✎	▁▁▂▁▃▄█▆▄▅▂▃ beats 38/39	very hard · avg 22%	▇ 1.9%	blunt · at its level
DSEffi-Bench Pass (1,000 data-science tasks, greedy)	61.3% ✎	▅··▃▅·█·▃██▃ beats 13/15	hard · avg 24%	▇ 1.6%	at its level

11 more tests

test	result	vs the others in its run	difficulty	counts	why
EQ-Bench 3 (emotional intelligence, judge Elo)	1383.4 ✎	▁·▂▁▄▅▅▃█▄▃▁ beats 60/62	avg 1044	▇ 1.5%	typical
WildClawBench	51.6% ✎	▃·▅▅█·▅█▅▃▃▃ beats 15/17	very hard · avg 11%	▇ 1.4%	at its level
AgentWorldBench (open-ended, overall)	57.8% ✎	▄▄··▄▄·▄█▄██ beats 10/11	hard · avg 25%	▇ 1.3%	blunt · at its level
ARC-AGI-1 public evaluation set (ARC Prize model_baseline, pass@2)	96.8% ✎	▁·▇▃▁·█▇▁·▇▃ beats 26/27	very hard · avg 4.1%	▇ 1.1%	sharp · noisy
FORTE (AGI-Eval)	73.2% ✎	▄···█·····▄█ beats 1/5	medium · avg 37%	▇ 1.0%	at its level
SimpleQA Verified	47.0% ✎	▁▁▃▂▃▃▆▆█▃▂▅ beats 47/78	very hard · avg 8.0%	▇ 1.0%	at its level
MMMLU lab	91.1% ✎	▂▁▁▂▃▂▄▄▅██▃ beats 45/46	medium · avg 61%	▇ 0.9%	small run
HLE-Verified full set, text-only (Qwen Team runs)	46.8% ✎	█·█·····█··█ beats 3/3	very hard · avg 14%	▇ 0.8%	at its level
FinSheet-Bench accuracy (501 questions, 24 spreadsheets)	80.2% ✎	█···██·····█ beats 1/3	medium · avg 52%	▇ 0.6%	blunt
DiagnosticIQ industrial maintenance MCQ (macro accuracy, 4 options)	73.6% ✎	▁··▃▁▁▁▅█▅▁▅ beats 24/24	medium · avg 52%	▇ 0.5%	blunt · at its level
Vectara hallucination leaderboard (HHEM-2.3, factual consistency rate)	87.8% ✎	▁▂▁▂▅█▂▃▄▃▁▁ beats 21/98	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… 41%, MathArena ArXivMath / BrokenA… 12%, LMArena leaderboard dataset 10% and 24 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	10	▇▇▇▇▇▇▇▇▇▇▇▇ 41%	CC BY 4.0
MathArena ArXivMath / BrokenArXiv / ArXivLean model-output datasets	6	▇▇▇ 12%	CC BY-SA 4.0
LMArena leaderboard dataset	5	▇▇▇ 10%	CC BY 4.0
Anthropic system cards and model announcements (lab-reported scores) · lab-reported	6	▇▇ 8%	Factual vendor-reported score (not licensed content)
ARC Prize Foundation model_baseline outputs on ARC-AGI-1/2 public evaluation sets	2	▇ 4%	MIT
Agents' Last Exam — authors' own fixed-harness model comparison	1	▇ 2%	CC BY 4.0
meituan-longcat/LongCat-2.0 model card	3	▇ 2%	mit
Knowledge Index of Noah's Ark (arXiv:2606.05104v2)	1	▇ 2%	CC BY 4.0
τ²/τ³-bench leaderboard: Sierra's own evaluation runs	1	▇ 2%	MIT
E2E-SWE: Benchmarking LLMs on Building Working Codebases from Scratch (arXiv:2609.38335v1)	1	▇ 2%	CC BY 4.0
ContractBench: Can LLM Agents Preserve Observation Contracts? (arXiv:2605.17281v2)	1	▇ 2%	CC BY 4.0
DSEffi-Bench: Demystifying Large Language Models' Capability in Efficient Data Science Code Generation — authors' own model runs	1	▇ 2%	CC BY 4.0
EQ-Bench 3 canonical leaderboard results	1	▇ 1%	MIT
Vibe Code Bench (Vals AI, arXiv:2603.04601v3)	1	▇ 1%	CC-BY-4.0
WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation (arXiv:2605.10912v1)	1	▇ 1%	CC BY 4.0
Qwen/Qwen-AgentWorld-35B-A3B model card	1	▇ 1%	Apache-2.0
SWE Atlas: Benchmarking Coding Agents Beyond Issue Resolution (arXiv:2605.08366v1)	1	▇ 1%	CC BY 4.0
One Success Isn't Reliability: Thinkingbox, a Sandbox and Benchmark for Agents in Stateful Business Workflows (arXiv:2608.19741v4)	1	▇ 1%	CC BY 4.0
HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam — Qwen Team re-evaluation on revised Humanity's Last Exam	1	▇ 1%	CC BY 4.0
MathNet: a Global Multimodal Benchmark for Mathematical Reasoning and Retrieval — authors' own model runs	1	▇ 1%	CC BY 4.0
BEAR-Bench: A Bilingual Enterprise and Academic Reasoning Benchmark for Multimodal Models (arXiv:2608.17895v1)	1	▇ 1%	CC BY 4.0
EEGAgentBench: Benchmarking LLM Agents on Short- and Long-Horizon EEG Analysis (arXiv:2609.31632v1)	1	▇ 1%	CC BY 4.0
FinSheet-Bench: From Simple Lookups to Complex Reasoning, Where LLMs Break on Financial Spreadsheets — authors' own model runs	1	▇ 1%	CC BY 4.0
DiagnosticIQ: A Benchmark for LLM-Based Industrial Maintenance Action Recommendation from Symbolic Rules (arXiv:2605.08614v1)	1	▇ 0%	CC BY 4.0
Diagram-MMU: A Multi-Modal Benchmark for Scientific Diagrams (arXiv:2608.12262v1)	1	▇ 0%	CC BY 4.0
T1-Bench: Benchmarking Multi-Scenario Agents in Real-World Domains (arXiv:2606.11070v1)	1	▇ 0%	CC BY 4.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)	1	▇ 0%	Apache-2.0
prior · pull toward the middle		▇ 0%
weighted average		= 74.7