llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsClaude Opus 4.6

Anthropic model product

Claude Opus 4.6

Claude Opus 4.6 is Anthropic's most intelligent model, improving on its predecessor's coding skills with more careful planning, longer agentic task sustenance, more reliable operation in larger codebases, and better code review and debugging skills. First Opus-class model with 1M token context window (beta), 128K output tokens, and adaptive thinking. Features effort controls (low/medium/high/max) and context compaction for long-running tasks. State-of-the-art on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp. Pricing: $5/$25 per million tokens (input/output).

Updated Aug 10, 2026. Default version: Claude Opus 4.6

Compare
LLMBoard score69.2Claude Opus 4.6
Coverage80%7 benchmark families
Context window1MTokens
Official input price$5Anthropic API

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
Claude Opus 4.6
Released
Feb 5, 2026
Knowledge cutoff
May 31, 2025
Parameters
N/A
Context window
1M
Max output
128K
Inputs
image, pdf, text
Outputs
text
Open weights
No
License
Proprietary

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

Claude Opus 4.6 category scores

Benchmark results

Published benchmark records for the scored version Claude Opus 4.6.

27 rows
Columns

Show columns

Graphwalks parents >128k1.0118100.0%CAug 7, 2026
OpenRCA0.311100.0%CAug 7, 2026
Tau2 Retail0.9126100.0%CAug 7, 2026
Tau2 Telecom1.0135100.0%CAug 7, 2026
Vending-Bench 28017.614100.0%CAug 7, 2026
FigQA0.82350.0%CAug 7, 2026
DeepSearchQA0.93871.4%CAug 7, 2026
Finance Agent0.63871.4%CAug 7, 2026
MRCR v2 (8-needle)0.832190.0%CAug 7, 2026
OSWorld0.732089.5%CAug 7, 2026
ARC-AGI v20.751673.3%CAug 7, 2026
SWE-bench Multilingual0.853487.9%CAug 7, 2026
AIME 20251.0611495.6%CAug 7, 2026
CyberGym0.761150.0%CAug 7, 2026
Legal Agent Benchmark0.061358.3%BAug 7, 2026
MMMLU0.964989.6%CAug 7, 2026
SWE-Bench Verified0.8710494.2%CAug 7, 2026
FrontierSWE0.681550.0%BAug 7, 2026
LiveBench0.8113873.0%BAug 7, 2026
Graphwalks BFS >128k0.6142238.1%CAug 7, 2026
Humanity's Last Exam0.5149285.7%CAug 7, 2026
BrowseComp0.8155875.4%CAug 7, 2026
GPQA0.91723393.1%CAug 7, 2026
Terminal-Bench 2.00.7174966.7%CAug 7, 2026
MCP Atlas0.6233024.1%CAug 7, 2026
MMMU-Pro0.8236565.6%CAug 7, 2026
CharXiv-R0.8274743.5%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

100 rows
Columns

Show columns

textspanish11512.02,292N/AAug 6, 2026
text factualityspanish11498.51,652N/AAug 6, 2026
documentoverall21509.937,271N/AJul 30, 2026
textcoding21534.620,444N/AAug 6, 2026
textenglish21506.334,062N/AAug 6, 2026
texthard prompts english21527.122,719N/AAug 6, 2026
text factualitychinese21544.93,008N/AAug 6, 2026
text factualitycoding21548.018,187N/AAug 6, 2026
text factualityenglish21499.233,696N/AAug 6, 2026
text factualityindustry software and it services21533.027,523N/AAug 6, 2026
text factualitykorean21461.1683N/AAug 6, 2026
text style controlrussian21509.37,658N/AAug 6, 2026
text style controlspanish21502.92,292N/AAug 6, 2026
textexpert31538.07,107N/AAug 6, 2026
textindustry business and management and financial operations31499.114,549N/AAug 6, 2026
textindustry software and it services31527.428,695N/AAug 6, 2026
textlonger query31516.330,370N/AAug 6, 2026
textmulti turn31506.912,925N/AAug 6, 2026
text factualityexpert31537.65,590N/AAug 6, 2026
text factualityhard prompts31518.846,806N/AAug 6, 2026
text factualityhard prompts english31530.220,552N/AAug 6, 2026
text factualityindustry mathematical31503.52,966N/AAug 6, 2026
text factualitymath31497.72,860N/AAug 6, 2026
text factualitymulti turn31504.010,451N/AAug 6, 2026
text style controlhard prompts31526.947,107N/AAug 6, 2026
text style controlhard prompts english31530.322,719N/AAug 6, 2026
visiondiagram31339.26,670N/AAug 6, 2026
document style controloverall41494.737,271N/AJul 30, 2026
texthard prompts41522.747,107N/AAug 6, 2026
textindustry entertainment and sports and media41480.315,598N/AAug 6, 2026
textindustry mathematical41518.83,900N/AAug 6, 2026
textinstruction following41509.623,799N/AAug 6, 2026
textrussian41499.77,658N/AAug 6, 2026
text factualityoverall41487.472,794N/AAug 6, 2026
text factualityexclude ties41500.355,246N/AAug 6, 2026
text factualityindustry entertainment and sports and media41470.713,567N/AAug 6, 2026
text factualityinstruction following41497.221,621N/AAug 6, 2026
text factualitylonger query41510.228,329N/AAug 6, 2026
text factualityrussian41500.45,829N/AAug 6, 2026
text style controlcoding41547.020,444N/AAug 6, 2026
text style controlenglish41506.934,062N/AAug 6, 2026
text style controlexpert41535.47,107N/AAug 6, 2026
text style controlindustry business and management and financial operations41502.614,549N/AAug 6, 2026
text style controlindustry entertainment and sports and media41478.115,598N/AAug 6, 2026
text style controlindustry mathematical41516.13,900N/AAug 6, 2026
text style controlindustry software and it services41535.528,695N/AAug 6, 2026
text style controlindustry writing and literature and language41490.617,487N/AAug 6, 2026
text style controlinstruction following41499.123,799N/AAug 6, 2026
text style controllonger query41518.030,370N/AAug 6, 2026
vision style controlchinese41358.71,311N/AAug 6, 2026
vision style controldiagram41330.66,670N/AAug 6, 2026
vision style controlhumor41306.2877N/AAug 6, 2026
textoverall51497.173,158N/AAug 6, 2026
textchinese51549.43,973N/AAug 6, 2026
textexclude ties51512.355,526N/AAug 6, 2026
textindustry legal and government51507.45,819N/AAug 6, 2026
textindustry writing and literature and language51492.717,487N/AAug 6, 2026
text factualityindustry legal and government51513.34,451N/AAug 6, 2026
text factualityindustry life and physical and social science51511.89,650N/AAug 6, 2026
text factualityindustry medicine and healthcare51514.84,018N/AAug 6, 2026
text factualityindustry writing and literature and language51480.915,158N/AAug 6, 2026
text factualitypolish51470.7867N/AAug 6, 2026
text style controloverall51497.573,158N/AAug 6, 2026
text style controlexclude ties51516.055,526N/AAug 6, 2026
text style controlindustry medicine and healthcare51514.05,320N/AAug 6, 2026
visionchinese51380.11,311N/AAug 6, 2026
visionhumor51324.2877N/AAug 6, 2026
visionocr51324.317,878N/AAug 6, 2026
textindustry life and physical and social science61510.612,007N/AAug 6, 2026
textindustry medicine and healthcare61499.85,320N/AAug 6, 2026
textnon english61484.739,095N/AAug 6, 2026
text style controlchinese61548.53,973N/AAug 6, 2026
text style controlindustry life and physical and social science61518.512,007N/AAug 6, 2026
text style controlmulti turn61511.712,925N/AAug 6, 2026
vision style controlocr61310.017,878N/AAug 6, 2026
textkorean71459.41,319N/AAug 6, 2026
textpolish71488.01,370N/AAug 6, 2026
text factualityfrench71508.01,808N/AAug 6, 2026
text style controlindustry legal and government71506.05,819N/AAug 6, 2026
visionoverall71311.024,609N/AAug 6, 2026
visionenglish71315.510,665N/AAug 6, 2026
vision style controlenglish71292.510,665N/AAug 6, 2026
textcreative writing81482.412,085N/AAug 6, 2026
textfrench81498.92,376N/AAug 6, 2026
textmath81509.53,965N/AAug 6, 2026
text factualitygerman81450.7533N/AAug 6, 2026
text factualityindustry business and management and financial operations81493.612,082N/AAug 6, 2026
text factualityjapanese81469.6474N/AAug 6, 2026
text style controlcreative writing81477.812,085N/AAug 6, 2026
text style controlmath81504.93,965N/AAug 6, 2026
text style controlnon english81484.539,095N/AAug 6, 2026
visioncreative writing vision81319.61,422N/AAug 6, 2026
text factualitycreative writing91468.59,770N/AAug 6, 2026
text factualitynon english91474.038,818N/AAug 6, 2026
vision style controloverall91292.924,609N/AAug 6, 2026
vision style controlcreative writing vision91297.31,422N/AAug 6, 2026
textgerman101488.91,148N/AAug 6, 2026
text style controlgerman101494.11,148N/AAug 6, 2026
text style controlkorean101458.51,319N/AAug 6, 2026
textjapanese121477.7726N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
$5 input, $25 output per 1M
Official provider
Anthropic
Lowest third-party
From $5 input, $25 output per 1M via LLM Gateway
Tracked offerings
16
16 rows
Columns

Show columns

LLM Gatewayclaude-opus-4-6global$5$251MAug 7, 2026
OpenCode Zenclaude-opus-4-6global$5$251MAug 7, 2026
302.AIclaude-opus-4-6global$5$251MAug 7, 2026
Neonclaude-opus-4-6global$5$251MAug 7, 2026
FrogBotclaude-opus-4-6global$5$25200KAug 7, 2026
Pioneerclaude-opus-4-6global$5$251MAug 7, 2026
Abacusclaude-opus-4-6global$5$251MAug 7, 2026
Azure Cognitive Servicesclaude-opus-4-6global$5$251MAug 7, 2026
Aurikoclaude-opus-4-6global$5$251MAug 7, 2026
Azureclaude-opus-4-6global$5$251MAug 7, 2026
AIHubMixclaude-opus-4-6global$5$251MAug 7, 2026
FreeModelclaude-opus-4-6global$5$251MAug 7, 2026
Requestyclaude-opus-4-6global$5$251MAug 7, 2026
Anthropicclaude-opus-4-6global$5$251MAug 7, 2026
Jiekou.AIclaude-opus-4-6global$5$251MAug 7, 2026
Venice AIclaude-opus-4-6global$6$301MAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

Claude Opus 4.6 versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

Claude Opus 4.6Feb 5, 202669.2N/A1M128KNoProprietary

What is Claude Opus 4.6?

A concise description based on the published model registry.

Claude Opus 4.6 is Anthropic's most intelligent model, improving on its predecessor's coding skills with more careful planning, longer agentic task sustenance, more reliable operation in larger codebases, and better code review and debugging skills. First Opus-class model with 1M token context window (beta), 128K output tokens, and adaptive thinking. Features effort controls (low/medium/high/max) and context compaction for long-running tasks. State-of-the-art on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp. Pricing: $5/$25 per million tokens (input/output).

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

Claude Opus 4.6 vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

Claude Opus 4.6vsGPT-5Claude Opus 4.6vsLlama 3.1 405BClaude Opus 4.6vsDeepSeek-V2.5Claude Opus 4.6vsGPT-5.2Claude Opus 4.6vsMiMo V2.5 ProClaude Opus 4.6vsGemma 4 26B A4B

Models similar to Claude Opus 4.6

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#22+4.2
AN

Claude Sonnet 3.5

Anthropic

73.4 LLMBoard

DetailsCompare
#41-5.6
AN

Claude Sonnet 3.7

Anthropic

63.6 LLMBoard

DetailsCompare
#13+9.2
AN

Claude Opus 4.7

Anthropic

78.4 LLMBoard

DetailsCompare
#55-10.4
AN

Claude Opus 3

Anthropic

58.8 LLMBoard

DetailsCompare
#6+13.6
AN

Claude Opus 4.8

Anthropic

82.9 LLMBoard

DetailsCompare
#110-28.0
AN

Claude Haiku 3.5

Anthropic

41.2 LLMBoard

DetailsCompare

FAQ

Common questions about Claude Opus 4.6.

When was Claude Opus 4.6 released?

Claude Opus 4.6's default version was released on Feb 5, 2026.

How much does Claude Opus 4.6 cost?

Claude Opus 4.6's official API price is $5 per million input tokens and $25 per million output tokens via Anthropic. The lowest tracked third-party offer starts at $5 input and $25 output via LLM Gateway.

Who created Claude Opus 4.6?

Claude Opus 4.6 is published under Anthropic in the model registry.

What is the context window for Claude Opus 4.6?

The default version has a 1M token context window.

Is Claude Opus 4.6 open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer Claude Opus 4.6?

16 published provider offerings are linked to the default version.

What models should I compare Claude Opus 4.6 with?

Nearby ranked alternatives include GPT-5, Llama 3.1 405B, DeepSeek-V2.5.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai