Sign Up

Claude Opus 4.5

Anthropic
Announcement
Claude Opus 4.5 launch post cover

Cost/1M tokens

input
$5.00
cached
$0.50
output
$25.00
cache write
$6.25
Context
200Ktokens
Released
11mo ago

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Arena leaderboards and other sourced evaluations appear below; missing scores are not zero.

How it ranks

Against its predecessor, similarly priced models and the models ranked around it.

DeepSWE 1.1

Best pass@1 per model · Higher is better

Not evaluated in DeepSWE 1.1

Arena Text

Best Arena rating per model · Higher is better

Arena #43

  1. 1511
    Muse WonderSimilar price
  2. 1485
    GPT-5.6 SolSimilar price
  3. 1481
    GPT-5.5Similar price
  4. 1474
    GLM 5.3 FlashNearby rank
  5. 1474
    GPT-6 AstraNearby rank
  6. 1474
    Claude Opus 4.5
  7. 1473
    GPT-5.5 InstantNearby rank
  8. 1472
    Claude Sonnet 5.5Nearby rank
  9. 1450
    Claude Opus 4.1Predecessor
  10. 1346
    GPT-4o (2024-05-13)Similar price

+24 vs Claude Opus 4.1 (predecessor) · Ahead of 1 of 4 similarly priced models

Arena WebDev

Best Arena rating per model · Higher is better

Arena #70

  1. 1649
    Muse WonderSimilar price
  2. 1615
    GPT-5.6 SolSimilar price
  3. 1512
    GPT-5.5Similar price
  4. 1506
    Hy3Nearby rank
  5. 1500
    Gemini 3.5 FlashNearby rank
  6. 1493
    Claude Opus 4.5
  7. 1484
    MiniMax M3Nearby rank
  8. 1482
    Qwen3.6 Max PreviewNearby rank
  9. 1462
    GPT-5.4Similar price
  10. 1389
    Claude Opus 4.1Predecessor

+104 vs Claude Opus 4.1 (predecessor) · Ahead of 1 of 4 similarly priced models

Arena Vision

Best Arena rating per model · Higher is better

Not ranked on Arena Vision yet

Arena Search

Best Arena rating per model · Higher is better

Not ranked on Arena Search yet

Arena Document

Best Arena rating per model · Higher is better

Arena #22

  1. 1486
    GPT-5.6 SolSimilar price
  2. 1481
    GPT-5.5Similar price
  3. 1474
    GPT-6 AstraSimilar price
  4. 1471
    GPT-5.4Similar price
  5. 1469
    Claude Sonnet 5Nearby rank
  6. 1468
    Gemini 3.7 FlashNearby rank
  7. 1467
    Claude Opus 4.5
  8. 1466
    Grok 4.5Nearby rank
  9. 1460
    GPT-5.6 LunaNearby rank

Ahead of 0 of 4 similarly priced models

Agent Arena

Best IPS score per model · Higher is better

Not ranked on Agent Arena yet

ARC-AGI-2

Best score per model · Higher is better

#45 of 85

  1. 95.0%
    GPT-6 AstraSimilar price
  2. 92.5%
    GPT-5.6 SolSimilar price
  3. 85.0%
    GPT-5.5Similar price
  4. 74.0%
    GPT-5.4Similar price
  5. 42.4%
    Qwen3.8 27BNearby rank
  6. 40.1%
    Inkling SmallNearby rank
  7. 37.6%
    Claude Opus 4.5
  8. 36.5%
    InklingNearby rank
  9. 31.1%
    Gemini 3 Pro PreviewNearby rank
  10. 8.6%
    Claude Opus 4Earlier version

+29.0 pts vs Claude Opus 4 (earlier version) · Ahead of 0 of 4 similarly priced models

FrontierMath-2025-02-28-Private

Best score per model · Higher is better

#24 of 58

  1. 47.6%
    GPT-5.4Similar price
  2. 47.2%
    Claude Opus 4.8Similar price
  3. 43.8%
    Claude Opus 4.7Similar price
  4. 40.7%
    Claude Opus 4.6Similar price
  5. 21.4%
    Kimi K2 ThinkingNearby rank
  6. 21.0%
    Qwen3.5 PlusNearby rank
  7. 20.7%
    Claude Opus 4.5
  8. 18.7%
    o3Nearby rank
  9. 16.4%
    GLM 5Nearby rank
  10. 7.2%
    Claude Opus 4.1Predecessor

+13.4 pts vs Claude Opus 4.1 (predecessor) · Ahead of 0 of 4 similarly priced models

FrontierMath-Tier-4-2025-07-01-Private

Best score per model · Higher is better

#18 of 46

  1. 31.3%
    Claude Opus 4.8Similar price
  2. 27.1%
    GPT-5.4Similar price
  3. 22.9%
    Claude Opus 4.7Similar price
  4. 22.9%
    Claude Opus 4.6Similar price
  5. 4.2%
    Claude Opus 4Nearby rank
  6. 4.2%
    Claude Opus 4.1Predecessor
  7. 4.2%
    Claude Opus 4.5
  8. 4.2%
    Claude Sonnet 4.5Nearby rank
  9. 4.2%
    Gemini 2.5 FlashNearby rank

Level with Claude Opus 4.1 (predecessor) · Ahead of 0 of 4 similarly priced models

FrontierMath-Tier-4-v2-Private

Best score per model · Higher is better

#61 of 72

  1. 97.6%
    GPT-6 AstraSimilar price
  2. 82.9%
    GPT-5.6 SolSimilar price
  3. 72.5%
    GPT-5.5Similar price
  4. 49.0%
    GPT-5.4Similar price
  5. 12.2%
    GPT-5.4 NanoNearby rank
  6. 9.8%
    GPT-5.4 MiniNearby rank
  7. 4.9%
    Claude Opus 4.5
  8. 4.9%
    InklingNearby rank
  9. 2.4%
    Claude Opus 4.1Predecessor

+2.4 pts vs Claude Opus 4.1 (predecessor) · Ahead of 0 of 4 similarly priced models

FrontierMath-Tiers-1-3-v2-Private

Best score per model · Higher is better

#64 of 87

  1. 93.7%
    GPT-6 AstraSimilar price
  2. 89.1%
    GPT-5.6 SolSimilar price
  3. 85.3%
    GPT-5.5Similar price
  4. 78.6%
    GPT-5.4Similar price
  5. 36.8%
    GLM 5.1Nearby rank
  6. 35.1%
    Qwen3.6 27BNearby rank
  7. 34.4%
    Claude Opus 4.5
  8. 34.4%
    Qwen3.7 PlusNearby rank
  9. 33.3%
    o3Nearby rank
  10. 12.6%
    Claude Opus 4.1Predecessor

+21.8 pts vs Claude Opus 4.1 (predecessor) · Ahead of 0 of 4 similarly priced models

GPQA diamond

Best score per model · Higher is better

#65 of 161

  1. 93.5%
    GPT-5.6 SolSimilar price
  2. 93.3%
    GPT-5.4Similar price
  3. 90.7%
    GPT-5.5Similar price
  4. 86.4%
    Qwen3.5 397B A17BNearby rank
  5. 86.2%
    GPT-5Nearby rank
  6. 86.0%
    Claude Opus 4.5
  7. 85.9%
    Claude Fable 5Nearby rank
  8. 85.9%
    Qwen3.6 27BNearby rank
  9. 77.3%
    Claude Opus 4.1Predecessor
  10. 48.9%
    GPT-4o (2024-05-13)Similar price

+8.8 pts vs Claude Opus 4.1 (predecessor) · Ahead of 1 of 4 similarly priced models

Humanity's Last Exam

Best score per model · Higher is better

#20 of 34

  1. 54.8%
    GPT-6 AstraSimilar price
  2. 36.2%
    GPT-5.4Similar price
  3. 36.2%
    Claude Opus 4.7Similar price
  4. 34.4%
    Claude Opus 4.6Similar price
  5. 20.3%
    o3Nearby rank
  6. 19.4%
    GPT-5 MiniNearby rank
  7. 14.2%
    Claude Opus 4.5
  8. 11.0%
    Gemini 2.5 Flash Preview 05-20Nearby rank
  9. 8.6%
    Gemini 3.1 Flash LiteNearby rank
  10. 7.9%
    Claude Opus 4.1Predecessor

+6.2 pts vs Claude Opus 4.1 (predecessor) · Ahead of 0 of 4 similarly priced models

OSWorld

Best score per model · Higher is better

#2 of 10

  1. 72.1%
    Claude Sonnet 4.6Similar price
  2. 66.3%
    Claude Opus 4.5
  3. 63.3%
    Kimi K2.5Nearby rank
  4. 62.9%
    Claude Sonnet 4.5Similar price
  5. 43.9%
    Claude Sonnet 4Earlier version

+22.4 pts vs Claude Sonnet 4 (earlier version) · Ahead of 1 of 2 similarly priced models

OTIS Mock AIME 2024-2025

Best score per model · Higher is better

#73 of 158

  1. 100.0%
    GPT-5.6 SolSimilar price
  2. 97.8%
    GPT-5.4Similar price
  3. 86.7%
    Qwen3.7 FlashNearby rank
  4. 86.4%
    GLM 5.2Nearby rank
  5. 86.1%
    Claude Opus 4.5
  6. 85.8%
    Claude Sonnet 4.6Nearby rank
  7. 84.7%
    Gemini 2.5 ProNearby rank
  8. 84.4%
    GPT-5.5Similar price
  9. 68.9%
    Claude Opus 4.1Predecessor
  10. 6.3%
    GPT-4o (2024-05-13)Similar price

+17.2 pts vs Claude Opus 4.1 (predecessor) · Ahead of 2 of 4 similarly priced models

SimpleQA Verified

Best score per model · Higher is better

#42 of 87

  1. 75.6%
    GPT-6 AstraSimilar price
  2. 69.7%
    GPT-5.6 SolSimilar price
  3. 63.0%
    GPT-5.5Similar price
  4. 46.3%
    GPT-5.4 ProNearby rank
  5. 45.8%
    Qwen3.8 MaxNearby rank
  6. 45.7%
    Claude Opus 4.5
  7. 45.1%
    GPT-5.4Similar price
  8. 44.1%
    Qwen3.6 PlusNearby rank

Ahead of 1 of 4 similarly priced models

SWE-bench Verified

Best score per model · Higher is better

#7 of 35

  1. 83.5%
    Claude Opus 4.7Similar price
  2. 76.9%
    GPT-5.4Similar price
  3. 76.7%
    Qwen3.6 Max PreviewNearby rank
  4. 76.7%
    Claude Opus 4.5
  5. 76.7%
    Kimi K2.6Nearby rank
  6. 75.6%
    Claude Opus 4.6Similar price
  7. 75.2%
    Claude Sonnet 4.6Similar price
  8. 73.3%
    Claude Opus 4.1Predecessor

+3.3 pts vs Claude Opus 4.1 (predecessor) · Ahead of 2 of 4 similarly priced models

Terminal Bench

Best score per model · Higher is better

#11 of 49

  1. 84.7%
    GPT-5.5Similar price
  2. 81.8%
    GPT-5.4Similar price
  3. 80.2%
    Claude Opus 4.7Similar price
  4. 79.8%
    Claude Opus 4.6Similar price
  5. 64.9%
    GPT-5.2Nearby rank
  6. 64.3%
    Gemini 3 Flash PreviewNearby rank
  7. 63.1%
    Claude Opus 4.5
  8. 61.6%
    GPT-5.1-Codex-MiniNearby rank
  9. 60.4%
    GPT-5.1-Codex-MaxNearby rank
  10. 38.0%
    Claude Opus 4.1Predecessor

+25.1 pts vs Claude Opus 4.1 (predecessor) · Ahead of 0 of 4 similarly priced models

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

All published scores

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.