Sign Up

Claude Sonnet 5.5

Anthropic
1Announcement
Claude Sonnet 5.5 launch post cover

Cost/1M tokens

input
$2.00
cached
$0.10
output
$10.00
cache write
$2.50
Context
1Mtokens
Released
1mo 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 #46

  1. 1526
    Gemini 4 CarbonSimilar price
  2. 1487
    Kimi K3Similar price
  3. 1482
    GPT-6.1 SolSimilar price
  4. 1474
    Claude Opus 4.5Nearby rank
  5. 1473
    GPT-5.5 InstantNearby rank
  6. 1472
    Claude Sonnet 5.5
  7. 1472
    Claude Sonnet 4.6Nearby rank
  8. 1468
    GPT-5.6 TerraNearby rank
  9. 1461
    Claude Sonnet 5Predecessor
  10. 1354
    Command ASimilar price

+12 vs Claude Sonnet 5 (predecessor) · Ahead of 1 of 4 similarly priced models

Arena WebDev

Best Arena rating per model · Higher is better

Arena #3

  1. 1836
    Claude Fable 5.2Nearby rank
  2. 1814
    Claude Opus 5.5Nearby rank
  3. 1773
    Claude Sonnet 5.5
  4. 1768
    Gemini 4 CarbonSimilar price
  5. 1764
    GPT-6 AstraNearby rank
  6. 1743
    GPT-6.1 SolSimilar price
  7. 1676
    Qwen3.8 Max (0902)Similar price
  8. 1658
    Kimi K3Similar price
  9. 1539
    Claude Sonnet 5Predecessor

+234 vs Claude Sonnet 5 (predecessor) · Ahead of 3 of 4 similarly priced models

Arena Vision

Best Arena rating per model · Higher is better

Arena #43

  1. 1293
    GPT-6.1 SolSimilar price
  2. 1280
    Grok 4.5Similar price
  3. 1280
    Gemini 3.1 Pro PreviewSimilar price
  4. 1263
    Grok 4.6Nearby rank
  5. 1263
    DeepSeek V4.1 FlashNearby rank
  6. 1263
    Claude Sonnet 5.5
  7. 1262
    Qwen3.7 PlusNearby rank
  8. 1261
    Claude Sonnet 5Predecessor
  9. 1199
    Mistral Medium 3.5Similar price

Level with Claude Sonnet 5 (predecessor, overlapping 95% CIs) · Ahead of 1 of 4 similarly priced models

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

Not ranked on Arena Document yet

Agent Arena

Best IPS score per model · Higher is better

Arena #4

  1. +0.195
    Claude Opus 5.5Nearby rank
  2. +0.170
    Gemini 4 CarbonSimilar price
  3. +0.160
    Claude Sonnet 5.5
  4. +0.148
    Claude Fable 5.1Nearby rank
  5. +0.145
    GPT-6 AstraNearby rank
  6. +0.125
    GPT-6.1 SolSimilar price
  7. +0.026
    Grok 4.7Similar price
  8. +0.020
    Claude Sonnet 5Predecessor
  9. +0.006
    Kimi K3Similar price

+0.140 vs Claude Sonnet 5 (predecessor) · Ahead of 2 of 4 similarly priced models

ARC-AGI-2

Best score per model · Higher is better

#17 of 85

  1. 94.2%
    Gemini 4 CarbonSimilar price
  2. 94.2%
    GPT-6.1 SolSimilar price
  3. 84.6%
    GPT-5.5 ProNearby rank
  4. 83.9%
    GPT-5.6 TerraNearby rank
  5. 83.9%
    Claude Sonnet 5.5
  6. 83.3%
    GPT-5.4 ProNearby rank
  7. 81.7%
    Inkling 2 SmallNearby rank
  8. 67.1%
    Grok 4.6Similar price
  9. 60.4%
    Claude Sonnet 4.6Earlier version
  10. 60.4%
    Kimi K3Similar price

+23.5 pts vs Claude Sonnet 4.6 (earlier version) · Ahead of 2 of 4 similarly priced models

FrontierMath-Tier-4-v2-Private

Best score per model · Higher is better

#11 of 72

  1. 100.0%
    GPT-6.1 SolSimilar price
  2. 95.1%
    Gemini 4 CarbonSimilar price
  3. 87.8%
    Claude Fable 5.1Nearby rank
  4. 82.9%
    GPT-5.6 SolNearby rank
  5. 80.5%
    Claude Sonnet 5.5
  6. 78.0%
    Gemini 4 ArgonNearby rank
  7. 78.0%
    GPT-5.5 ProNearby rank
  8. 39.0%
    Kimi K3Similar price
  9. 34.1%
    Qwen3.8 Max (0902)Similar price
  10. 29.3%
    Claude Sonnet 5Predecessor

+51.2 pts vs Claude Sonnet 5 (predecessor) · Ahead of 2 of 4 similarly priced models

FrontierMath-Tiers-1-3-v2-Private

Best score per model · Higher is better

#10 of 87

  1. 93.7%
    GPT-6.1 SolSimilar price
  2. 90.5%
    Gemini 4 CarbonSimilar price
  3. 89.8%
    GPT-6 SolNearby rank
  4. 89.1%
    GPT-5.6 SolNearby rank
  5. 88.8%
    Claude Sonnet 5.5
  6. 87.7%
    GPT-5.5 ProNearby rank
  7. 87.0%
    Claude Fable 5Nearby rank
  8. 72.2%
    Kimi K3Similar price
  9. 66.0%
    Grok 4.6Similar price
  10. 65.6%
    Claude Sonnet 5Predecessor

+23.2 pts vs Claude Sonnet 5 (predecessor) · Ahead of 2 of 4 similarly priced models

GPQA diamond

Best score per model · Higher is better

#4 of 161

  1. 96.5%
    Gemini 4 CarbonSimilar price
  2. 96.4%
    Gemini 4 ArgonNearby rank
  3. 95.8%
    GPT-6 AstraNearby rank
  4. 95.6%
    Claude Sonnet 5.5
  5. 95.4%
    Gemini 3.8 FlashNearby rank
  6. 95.4%
    GPT-6.1 SolSimilar price
  7. 94.0%
    Grok 4.6Similar price
  8. 93.1%
    Kimi K3Similar price
  9. 90.5%
    Claude Sonnet 5Predecessor

+5.1 pts vs Claude Sonnet 5 (predecessor) · Ahead of 3 of 4 similarly priced models

Humanity's Last Exam

Best score per model · Higher is better

#4 of 34

  1. 46.5%
    Claude Fable 5.1Nearby rank
  2. 46.4%
    Gemini 3.1 Pro PreviewSimilar price
  3. 46.1%
    Claude Sonnet 5.5
  4. 44.5%
    Gemini 3.8 FlashNearby rank
  5. 44.3%
    GPT-5.4 ProNearby rank
  6. 20.3%
    o3Similar price
  7. 7.5%
    Claude Sonnet 4.5Earlier version
  8. 5.4%
    GPT-4.1Similar price
  9. 2.7%
    GPT-4o (2024-11-20)Similar price

+38.6 pts vs Claude Sonnet 4.5 (earlier version) · Ahead of 3 of 4 similarly priced models

OTIS Mock AIME 2024-2025

Best score per model · Higher is better

#1 of 158

  1. 100.0%
    Claude Fable 5.1Nearby rank
  2. 100.0%
    Claude Opus 5.5Nearby rank
  3. 100.0%
    Claude Sonnet 5.5
  4. 100.0%
    Gemini 4 ArgonSimilar price
  5. 100.0%
    Gemini 4 CarbonNearby rank
  6. 100.0%
    GPT-6 SolSimilar price
  7. 100.0%
    Qwen3.8 Max (0902)Similar price
  8. 97.2%
    Kimi K3Similar price
  9. 94.7%
    Claude Sonnet 5Predecessor

+5.3 pts vs Claude Sonnet 5 (predecessor) · Ahead of 1 of 4 similarly priced models

SimpleQA Verified

Best score per model · Higher is better

#39 of 87

  1. 77.7%
    Gemini 4 CarbonSimilar price
  2. 73.9%
    GPT-6.1 SolSimilar price
  3. 56.0%
    Grok 4.7Similar price
  4. 50.6%
    Kimi K3Similar price
  5. 47.0%
    Claude Opus 4.6Nearby rank
  6. 47.0%
    DeepSeek V4 Pro 0423Nearby rank
  7. 46.5%
    Claude Sonnet 5.5
  8. 46.3%
    GPT-5.4 ProNearby rank
  9. 45.8%
    Qwen3.8 MaxNearby rank
  10. 33.7%
    Claude Sonnet 5Predecessor

+12.8 pts vs Claude Sonnet 5 (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.

Reasoning effort

Arena WebDev

Score by reasoning effort · Higher is better

ARC-AGI-2

Score by reasoning effort · Higher is better

All published scores

  • Arena TextArena #461472

    +12 vs Claude Sonnet 5 (predecessor)

    • xhigh147295% CI 1466 to 1479 · 13,821 votes
    Source:Arena Text leaderboard, 2026-11-09
  • Arena WebDevArena #31773

    +234 vs Claude Sonnet 5 (predecessor)

    • high171595% CI 1707 to 1724 · 7,881 votes
    • xhigh177395% CI 1763 to 1782 · 5,818 votes
    Source:Arena WebDev leaderboard, 2026-11-10
  • Arena VisionArena #431263

    Level with Claude Sonnet 5 (predecessor, overlapping 95% CIs)

    • xhigh126395% CI 1253 to 1273 · 4,257 votes
    Source:Arena Vision leaderboard, 2026-11-09
  • Agent ArenaArena #4+0.160

    +0.140 vs Claude Sonnet 5 (predecessor)

    • max+0.16095% CI +0.131 to +0.188 · 1,274,623 observations
    Source:Agent Arena leaderboard, 2026-11-11
  • ARC-AGI-2#17 of 8583.9%

    +23.5 pts vs Claude Sonnet 4.6 (earlier version)

    • low24.0%
    • medium32.8%
    • high67.4%
    • xhigh83.9%
    Source:External evaluation via Epoch AIOriginal-source terms apply
  • FrontierMath-Tier-4-v2-Private#11 of 7280.5%

    +51.2 pts vs Claude Sonnet 5 (predecessor)

    • max80.5%
    Source:Epoch AI evaluation
  • FrontierMath-Tiers-1-3-v2-Private#10 of 8788.8%

    +23.2 pts vs Claude Sonnet 5 (predecessor)

    • max88.8%
    Source:Epoch AI evaluation
  • GPQA diamond#4 of 16195.6%

    +5.1 pts vs Claude Sonnet 5 (predecessor)

    • max95.6%
    Source:Epoch AI evaluation
  • Humanity's Last Exam#4 of 3446.1%

    +38.6 pts vs Claude Sonnet 4.5 (earlier version)

    • xhigh46.1%
    Source:External evaluation via Epoch AIOriginal-source terms apply
  • OTIS Mock AIME 2024-2025#1 of 158100.0%

    +5.3 pts vs Claude Sonnet 5 (predecessor)

    • max100.0%
    Source:Epoch AI evaluation
  • SimpleQA Verified#39 of 8746.5%

    +12.8 pts vs Claude Sonnet 5 (predecessor)

    • max46.5%
    Source:Epoch AI evaluation

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.