Anthropic, released Feb 5, 2026

Claude Opus 4.6price, context, benchmarks and release details

Provisional: not enough results to rank yet 100% confidence 100 percent, Full confidence, 8 of 9 expected sources in
70.8
SI Score
#10 of 142 ranked models
Input, per 1M tokens
$5.00Anthropic API pricingOfficial Claude API base tokens; lowest short-context global Standard rate; cache, batch, fast and regional premiums excluded Retrieved Oct 9, 2026 · factual citation
Open source ↗
Output, per 1M tokens
$25.00Anthropic API pricingOfficial Claude API base tokens; lowest short-context global Standard rate; cache, batch, fast and regional premiums excluded Retrieved Oct 9, 2026 · factual citation
Open source ↗
Context window
1Mmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Max output
128Kmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
Released
Feb 5, 2026models.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗

How the score breaks down

Coding (weight 40 percent) 62.4
Math (weight 15 percent) 74.1
Preference (weight 15 percent) 81.6
Reasoning (weight 30 percent) 78.8

Weights: reasoning 30%, math 15%, coding 40%, preference 15%. Results use fixed 0–100 scales before averaging, and thin evidence is pulled toward 50. Method si-v3-retained-evidence-2, computed Oct 9, 2026, 05:13 UTC.

Around it on the leaderboard

  1. 8 Claude Opus 4.7 71.2
  2. 9 Kimi K3 71.0
  3. 10 Claude Opus 4.6 70.8
  4. 11 Muse Spark 1.3 70.1
  5. 12 Gemini 3.7 Flash 70.0

Full leaderboard

Benchmark results

28 benchmarks, 46 results

Each row shows the best published result. Where a model was tested at several settings, such as reasoning effort, open the row to see each one. Hover or tap a value for its source.

ARC-AGI-1 (public eval)reasoning 96.8%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'max' output effort. [variant] claude-opus-4-6-thinking-120K-maxPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 96.8 4 settings
  • Setting 1 96.3%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 2 89.6%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'low' output effort. [variant] claude-opus-4-6-thinking-120K-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 3 96.8%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'max' output effort. [variant] claude-opus-4-6-thinking-120K-maxPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 4 94.8%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'medium' output effort. [variant] claude-opus-4-6-thinking-120K-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
About ARC-AGI-1 (public eval)
ARC-AGI-1 (semi-private)reasoning 94%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 94.0 4 settings
  • Setting 1 94%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 2 86%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'low' output effort. [variant] claude-opus-4-6-thinking-120K-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 3 93%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'max' output effort. [variant] claude-opus-4-6-thinking-120K-maxPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 4 92%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'medium' output effort. [variant] claude-opus-4-6-thinking-120K-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
About ARC-AGI-1 (semi-private)
ARC-AGI-2 (public eval)reasoning 79.0%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 79.0 4 settings
  • Setting 1 79.0%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 2 59.9%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'low' output effort. [variant] claude-opus-4-6-thinking-120K-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 3 74.9%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'max' output effort. [variant] claude-opus-4-6-thinking-120K-maxPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 4 73.6%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'medium' output effort. [variant] claude-opus-4-6-thinking-120K-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
About ARC-AGI-2 (public eval)
ARC-AGI-2 (semi-private)reasoning 69.2%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 69.2 4 settings
  • Setting 1 69.2%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'high' output effort. [variant] claude-opus-4-6-thinking-120K-highPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 2 64.6%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'low' output effort. [variant] claude-opus-4-6-thinking-120K-lowPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 3 68.8%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'max' output effort. [variant] claude-opus-4-6-thinking-120K-maxPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
  • Setting 4 66.3%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. Evaluation conducted with a 120K thinking budget and a maximum context window of 128K tokens and 'medium' output effort. [variant] claude-opus-4-6-thinking-120K-mediumPublished Oct 6, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
About ARC-AGI-2 (semi-private)
GPQA Diamondreasoning 90.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Feb 6, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 90.5 3 settings
  • max effort 88.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 6, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
  • 32K thinking 90.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Feb 6, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
  • 64K thinking 88.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 64KPublished Feb 6, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
About GPQA Diamond
Humanity's Last Exam (Scale AI)reasoning 19%Humanity’s Last ExamPotential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions. [variant] Published Feb 17, 2026 Retrieved Oct 9, 2026 · factual citation
Open source ↗ 19.0
LiveBench Reasoning: connectionsreasoning 99.3%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 99.3
LiveBench Reasoning: consecutive eventsreasoning 63.5%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 63.5
LiveBench Reasoning: logic with navigationreasoning 80%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 80.0
LiveBench Reasoning: spatialreasoning 98%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 98.0
LiveBench Reasoning: theory of mindreasoning 82.7%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 82.7
LiveBench Reasoning: zebra puzzlesreasoning 94%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 94.0
FrontierMath Tier 4 (v2)math 26.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 26.8
FrontierMath Tiers 1–3 (v2)math 66.0%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Jun 11, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 66.0
LiveBench Math: AMPS Hardmath 97%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 97.0
LiveBench Math: competition mathmath 95.1%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 95.1
LiveBench Math: integralsmath 73%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 73.0
LiveBench Math: olympiadmath 92.2%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 92.2
LiveBench Math: simplifymath 59.1%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 59.1
OTIS Mock AIME 2024–2025math 94.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 64KPublished Feb 6, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 94.4 3 settings
  • max effort 91.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] maxPublished Aug 6, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
  • 32K thinking 93.1%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 32KPublished Feb 6, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
  • 64K thinking 94.4%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] 64KPublished Feb 6, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
About OTIS Mock AIME 2024–2025
LiveBench Coding: code completioncoding 76.1%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 76.1
LiveBench Coding: code generationcoding 80.3%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 80.3
LiveBench Coding: JavaScriptcoding 63.6%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 63.6
LiveBench Coding: Pythoncoding 50%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 50.0
LiveBench Coding: TypeScriptcoding 33.3%LiveBenchLiveBench subtask result; dataset edition 2026_06_25; publication time is HTTP Last-Modified of the score CSV, not the dataset edition or individual evaluation date [variant] 2026_06_25Published Oct 7, 2026 Retrieved Oct 9, 2026 · factual citation; Apache-2.0 code
Open source ↗ 33.3
SWE-bench Pro (public)coding 51.9%Official model cards via models.devLab-reported; metric resolve rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] public Retrieved Oct 9, 2026 · factual citation; MIT transcription
Open source ↗ 51.9 2 settings
  • Setting 1 51.9%Official model cards via models.devLab-reported; metric resolve rate; transcribed by MIT models.dev catalog; not independently evaluated [variant] public Retrieved Oct 9, 2026 · factual citation; MIT transcription
    Open source ↗
  • Setting 2 47.1%SWE-bench Pro (public)Published steward score [variant] Published Apr 8, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
About SWE-bench Pro (public)
SWE-bench Verifiedcoding 78.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Feb 18, 2026 Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 78.7 2 settings
  • Setting 1 78.7%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] Published Feb 18, 2026 Retrieved Oct 9, 2026 · CC-BY
    Open source ↗
  • Setting 2 75.6%SWE-bench VerifiedSWE-bench published model plus agent result; harness retained, not a base model evaluation [variant] mini-SWE-agent; 2.0.0Published Feb 17, 2026 Retrieved Oct 9, 2026 · factual citation
    Open source ↗
About SWE-bench Verified
LMArena Textpreference 1498.1 eloLMArena / ArenaPublished source fact [variant] text / overallPublished Oct 8, 2026 Retrieved Oct 9, 2026 · CC-BY-4.0
Open source ↗ 81.6

Normalization uses a fixed 0–100 scale for each unit, independent of other models. Compare evaluation conditions before reading a small gap as decisive. “Lab-reported” marks the provider's own published figure.

Details and sources

Open weights
Nomodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
License
not yet reported
Input modalities
text, image, pdfmodels.devPublished source fact Retrieved Oct 9, 2026 · MIT
Open source ↗
First seen by SuperIndex
Oct 8, 2026
Coverage
95% of expected source weight

Reported (8)

  • ARC PrizeOct 8, 2026
  • Epoch AI BenchmarkingOct 8, 2026
  • Humanity’s Last ExamOct 8, 2026
  • Official model cards via models.devOct 8, 2026
  • LiveBenchOct 8, 2026
  • LMArena / ArenaOct 8, 2026
  • SWE-bench VerifiedOct 8, 2026
  • SWE-bench Pro (public)Oct 8, 2026

Awaiting (1)

  • Terminal-Bench5% of weight

Confidence rises as pending sources publish. Some sources never cover some models, so confidence reaches 100% at 80% of expected weight.

What changed

All releases · RSS feed

Alerts on this device

What to be alerted about
RSS feed