Inference
EN
Anthropic

Claude Opus 4.6

Claude Opus 4.6 is a standard model from Anthropic. 42 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Standard
  • Input: text · image · file
  • Released 2026-02
RankedEnough results for an index.
IndexAll categories, out of 100.58.0±3.0Median 51.0
CoverageIndex weight that has results.95%
SpeedOutput tokens per second.36/sMedian 67/s
First tokenSeconds to the first output token.1.1 sMedian 1.3 s
Input / 1MUS dollars per 1M input tokens.$5.00batch $2.50Median $0.60
Output / 1MUS dollars per 1M output tokens.$25.00batch $12.50 Batch price per 1M output tokens.Median $2.55
ContextMaximum tokens in one request.1MMedian 262K
EloLMArena rating and rank.1497±3 #10Median 1422

Details

From OpenRouter
API IDThe model ID on OpenRouter.
anthropic/claude-opus-4.6
Max outputMaximum output tokens per request.
128K
Knowledge cutoffLast date of training data.
—
ToolsTool calls on OpenRouter.
Yes
JSON outputOutput follows a given JSON schema.
Yes
Cache read / 1MUS dollars per 1M cached tokens.
$0.50
WeightsLink to the published weights.
—
RetiresDate OpenRouter removes the model.
—

Where to run

APIs that serve this model.
8 providers
ProviderTypeMaker, cloud, host or router.InputOutputContext
Amazon BedrockCloud$5.00$25.001M
DatabricksCloud$5.00$25.001M
DigitalOceanCloud$5.00$25.00200K
GMI CloudHost$5.00$25.00410K
Google Vertex AICloud$5.00$25.001M
OpenRouterRouter$5.00$25.001M
SAP AI CoreCloud$5.00$25.001M
Snowflake CortexCloud$5.00$25.001M

US dollars per 1M tokens, list price. From models.dev (opens in a new tab) (MIT).

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.12 counted57.883 / 218
CodingCode writing and repair.6 counted59.274 / 209
ReasoningLogic problems and puzzles.7 counted55.385 / 227
MultimodalTasks with images and text.4 counted52.663 / 84
KnowledgeFacts and expert knowledge.7 counted61.856 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted30.7—
MathMath problems.5 counted63.837 / 168

Results

42 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
SuperGPQA: Scaling LLM Evaluation Across 285 Graduate DisciplinesKnowledge95.0%1 / 22
SWE-bench Verified · SWE-bench/experimentsCoding75.6%5 / 65
FrontierMath-2025-02-28-PrivateMath40.7% ±2.97 / 69
Software Engineering Benchmark VerifiedCoding80.8%14 / 114
FrontierMath-Tier-4-2025-07-01-PrivateMath22.9% ±6.18 / 55
Graduate-Level Google-Proof Q&AKnowledge91.3%22 / 141
MMLU-Pro first-party comparison snapshotKnowledge89.1%1 / 6
Gert Labs Composite Game BenchmarkAgentic61.9%10 / 52
SWE-bench Verified (mini-swe-agent-v2)Coding75.6%1 / 5
Artificial Analysis Omniscience AccuracyKnowledge45.8%34 / 164
Claw-EvalAgentic70.4%9 / 42
SWE-Bench verified · Epoch AICoding77.2% ±2.07 / 32
Humanity's Last ExamKnowledgeAbove fitFar above this model's index.53.0%17 / 77
GPQA diamond · Epoch AIKnowledge88.4% ±2.354 / 217
OTIS Mock AIME 2024-2025Math91.1% ±4.358 / 200
Terminal-Bench 2.0 · BenchLMAgentic65.4%12 / 41
React Native EvalsCoding84.1%5 / 16
BrowseCompAgentic83.7%21 / 63
FrontierMath-Tiers-1-3-v2-PrivateMath66.0% ±2.835 / 90
τ²-Bench Tool-Agent-User EvaluationAgentic84.8%58 / 149
ScreenSpot ProMultimodal83.1%8 / 20
SimpleQA VerifiedKnowledge47.0% ±1.637 / 91
Artificial Analysis GPQA DiamondKnowledge84.0%86 / 199
Massive Multitask Language Understanding ProfessionalKnowledge82.0%33 / 75
GPQA Diamond · BenchLMKnowledge89.2%37 / 84
Mystery Game PuzzlesReasoning25.0% ±4.437 / 80
Critical Physics TasksReasoning2.8%101 / 214
Terminal-Bench HardAgentic48.5%11 / 23
Vibe Code Bench v1.1Coding57.6% ±4.461 / 123
Artificial Analysis Humanity's Last ExamKnowledge19.1%116 / 222
Humanity's Last Exam without toolsKnowledge40.0%28 / 53
FrontierMath-Tier-4-v2-PrivateMath26.8% ±7.039 / 72
Massive Multi-discipline Multimodal Understanding ProMultimodal77.3%29 / 53
OSWorld-VerifiedAgentic72.7%21 / 37
τ²-bench BankingAgentic27.3%16 / 28
JobBenchAgentic36.7%21 / 36
Artificial Analysis IFBenchInstruction44.6%87 / 146
Artificial Analysis Long Context ReasoningReasoning67.0%130 / 217
Artificial Analysis MMMU-ProMultimodal72.5%69 / 113
ARC-AGI-3 (semi-private)Reasoning0.5%16 / 26
Chess PuzzlesReasoning21.0% ±4.159 / 90
MirrorCodeCoding33.3% ±10.026 / 39
SWE-bench ProCoding53.4%61 / 83
Furniture AssemblyReasoning28.3% ±5.928 / 37
EBR-benchReasoning12.7%24 / 31
DeepSearchQAAgentic73.7%19 / 24
ParseBench: document parsing for AI agents (LlamaIndex leaderboard)Multimodal53.8%32 / 39
CyberGymAgentic66.6%24 / 29
ERQAMultimodal51.6%16 / 18
FrontierCode 1.1 MainCoding26.9%16 / 17
ApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable jobAgentic5.0%17 / 18
AndroidWorldAgentic62.0%10 / 10
HealthBench HardKnowledge14.8%13 / 13
FrontierMath-2025-02-28-PublicMath90.0% ±10.02 / 64
FrontierMath-Tier-4-2025-07-01-PublicMath0.0%3 / 36
SWE-bench Multilingual · SWE-bench/experimentsCoding72.0%2 / 14
MedScribeHealthcare86.7% ±1.921 / 109
MedCodeHealthcare48.2% ±2.030 / 106
AIME25 first-party comparison snapshotMath99.8%—
LiveCodeBench ProCoding70.7%—
SWE-RebenchCoding65.3%—
MedXpertQA MultimodalMultimodal64.8%—
MedXpertQA TextKnowledge52.1%—
ResearchClawBenchAgentic19.9%—

Sources

BenchLMCC BY-NC 4.0OpenRouterTerms restrict reuseEpoch AICC BYSWE-bench teamNo licence statedVals AINo licence statedSierra ResearchMITARC Prize FoundationTerms restrict reusemodels.devMIT

History

5 releases
Index58.8 → 58.0
23 Oct 202612 Nov 2026
Output / 1M$25.00 → $25.00
23 Oct 202612 Nov 2026

Questions

How much does Claude Opus 4.6 cost?
$5.00 per 1M input tokens, $25.00 per 1M output tokens.
How good is Claude Opus 4.6?
Its index is 58.0 out of 100. That is place 84 of 283 ranked models.
What is Claude Opus 4.6 best at?
Knowledge: 61.8 out of 100, place 56 of 268.
How fast is Claude Opus 4.6?
36 output tokens per second, 1.1 s to the first token.
What is the context window of Claude Opus 4.6?
1M tokens.
Is Claude Opus 4.6 open source?
No. Its weights are not published.
When was Claude Opus 4.6 released?
February 2026.

Badge

Claude Opus 4.6 on Inference360: place 84 of 283, index 58.0
HTML
<a href="https://inference360.ai/models/claude-opus-4-6"><img src="https://inference360.ai/badges/claude-opus-4-6.svg" alt="Claude Opus 4.6 on Inference360: place 84 of 283, index 58.0" height="20"></a>
Markdown
[![Claude Opus 4.6 on Inference360: place 84 of 283, index 58.0](https://inference360.ai/badges/claude-opus-4-6.svg)](https://inference360.ai/models/claude-opus-4-6)

Compare with

Gemini 3 Pro57.9Kimi K2.658.2Kimi K2.7 Code57.7Beam 1.0 501B-A23B NVFP458.5Claude Sonnet 4.657.0

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeGLM-5.3-Flash63.1 · $0.078/1M

More from Anthropic

All models
Claude Opus 4.6 (Adaptive)60.0Claude Sonnet 4.657.0Claude Mythos PreviewUnrankedClaude Opus 4.5 Thinking56.3Claude Opus 4.552.0