The Model Index
Models/Anthropic

Claude Sonnet 5.5

AnthropicClosed weightsFlagship

Claude Sonnet 5.5 is a closed AI model from Anthropic with an index score of 187 (GPT-4 = 100), rank 12 of 217.

187171–203Index score#12 of 217 models
$2.0/ $10Price per million tokensinput / output · via OpenRouter
1MContext windowtokens it reads at once
28 Sep 2026Released2 months ago

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100Claude Sonnet 5.5

Claude Fable 5.1 reached this score 1 month earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
96%
Humanity's Last Exam
46%
Mathsevidence
FrontierMath (tiers 1-3)
89%
Codingevidence
SWE-bench Verified
no result
Terminal-Bench 4.0
64%
Agentsevidence
APEX-Agents
76%
Professional workevidence
Finance Agent (v2)
58%
Novel problemsevidence
ARC-AGI-2
84%
Long tasksno evidence yet
METR time horizon
no result

Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.

What the lab says1 of 8 also run by third parties

TestAnthropic saysThird parties
Terminal-Bench 4.070.6%64.1%Vals AI
FrontierCode 1.1 (Main)max effort46.2%not run yet
CursorBench 4.055.5%not run yet
GDPval-AA v2.11844 Elonot run yet
AA-Briefcase v1.11811 Elonot run yet
Humanity's Last Examwith tools64.5%not run yet
OSWorld 2.1partial credit80.1%not run yet
Chartographyno tools61.6%not run yet

From Anthropic’s launch post, Introducing Claude Sonnet 5.5, copied from the table in the post. Shown, never scored; settings can differ from the third-party run.

What others sayrank on 11 leaderboards

Best attop three among current models

Where to use itprices checked 18 Nov 2026

Host prices from OpenRouter; going direct can differ.

Facts and sources

Released28 Sep 2026
Announcementanthropic.com ↗
Release entryClaude Sonnet 5.5
Scored asClaude Sonnet 5.5
Output speed–
Hugging Faceclosed weights

Anthropic releasesaround this one

This page as a chartDownload PNG