HolyGrAIL

All tasks / Models

Claude Sonnet 5.5

Anthropic's mid-tier Claude model, described as its best combination of speed and intelligence.

  • Anthropic
  • Closed
  • Released 28 Sep 2026

Official page

Where it ranks

  • A
    Chat & assistants
    Models · Cloud · 1,475 on LMArena Text Arena (overall, style control)
  • –
    Search & research
    Models · Cloud · no result on this task’s benchmarks yet
  • A
    Writing
    Models · Cloud · 1,465 on LMArena Text Arena: Creative Writing (style control)
  • –
    Documents & OCR
    Models · Cloud · no result on this task’s benchmarks yet
  • S
    Coding
    Models · Cloud · 1,768 on LMArena WebDev / Code Arena (overall)
  • A
    Image understanding
    Models · Cloud · 1,266 on LMArena Vision Arena (overall, style control)
  • –
    Browser & computer agents
    Models · Cloud · no result on this task’s benchmarks yet
  • A
    Workflow automation
    Models · Cloud · 71.8% on AutomationBench-AA (Artificial Analysis)

Facts

Vendor
Anthropic
Family
Claude Sonnet
Released
28 Sep 2026
Weights
Closed: API and apps only
License
proprietary
API price
$2 in / $10 out per 1M tokens
Access
Paid only
Last verified
18 Oct 2026

Price from platform.claude.com, 18 Oct 2026.

Benchmark results

BenchmarkResultEntry on the leaderboardAs ofSource
Artificial Analysis Intelligence Index56Claude Sonnet 5.5 (Max, Default Fallback)
Raw value from the page's embedded model data; the site displays it rounded to an integer.
18 Oct 2026artificialanalysis.ai
AutomationBench-AA (Artificial Analysis)71.8%Claude Sonnet 5.5 (Max, Default Fallback)
AutomationBench-AA Score (share of each task's objectives completed, a task with any guardrail violation scoring zero); raw value from the page's embedded model data (displayed rounded); rank 3 of 32; Tasks Completed 45.51%, Objectives Completed 90.97%; no last-updated date shown, so as_of = retrieval date. 'Default Fallback' means Anthropic's classifier fallback to another model was left on, as in AA's default runs.
19 Oct 2026artificialanalysis.ai
LMArena Text Arena (overall, style control)1,475claude-sonnet-5.5-xhigh
From LMArena's official leaderboard-dataset (text_style_control, split latest, category overall, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text (site blocks automated fetches).
15 Oct 2026huggingface.co
LMArena Text Arena: Creative Writing (style control)1,465claude-sonnet-5.5-xhigh
From LMArena's official leaderboard-dataset (text_style_control, split latest, category creative_writing, published 2026-10-15), which mirrors https://arena.ai/leaderboard/text/creative-writing (site blocks automated fetches). Rank 22 of 414; 1532 votes; 95% CI 1448.9-1481.3.
15 Oct 2026huggingface.co
LMArena Vision Arena (overall, style control)1,266claude-sonnet-5.5-xhigh
From LMArena's official leaderboard-dataset (vision_style_control, split latest, category overall, published 2026-10-09), which mirrors https://arena.ai/leaderboard/vision (site blocks automated fetches). Rank 39 of 165; 2068 votes; 95% CI 1252.5-1279.9.
9 Oct 2026huggingface.co
LMArena WebDev / Code Arena (overall)1,768Claude Sonnet 5.5 (xHigh)
From LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 3; 3342 votes; 95% CI 1754.9-1780.7.
18 Oct 2026huggingface.co
ParseBench (models)70.2%Anthropic Sonnet 5.5
Overall (mean of tables, charts, content faithfulness, semantic formatting, visual grounding) from leaderboard.csv in the ParseBench repo (last leaderboard commit 2026-10-09).
9 Oct 2026github.com
SWE-Bench Pro V2 (public set, full)95.8%Sonnet 5.5 (mini-swe-agent) xhigh
SWE-Bench Pro V2 Full (642 tasks); row added 2026-10-12; 95% CI +/-1.3; harness in the variant name; read from the leaderboard page's embedded data.
19 Oct 2026labs.scale.com

Where you can use it

Earlier versions

  • Claude Sonnet 5 · 30 Jun 2026
  • Claude Sonnet 4.6 · 17 Feb 2026
  • Claude Sonnet 4.5 · 29 Sep 2025

Compare

Alternatives to Claude Sonnet 5.5

Sources