NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Claude Sonnet 5.5
Anthropic logo

Claude Sonnet 5.5

by Anthropic · Released Sep 2026

Multimodal1M Context
81.4
avg score
Rank #20
Compare
Better than 94% of all models
Context
1.0M tokens (~500 books)
Input $/1M
$2.00
Output $/1M
$10.00
Type
multimodal
License
Proprietary
Benchmarks
27 tested
Data as of
About

Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...

Tested on 27 benchmarks · BenchGecko score 81.4. Top scores: OTIS Mock AIME 2024-2025 (100.0%), Proofbench (100.0%), GPQA diamond (98.0%).

Looking for similar performance at lower cost?
MiMo-V2-Flash scores 81.7 (100% as good) at $0.09/1M input · 96% cheaper
Capabilities
coding
70.6
#19 globally
reasoning
83.8
#18 globally
math
91.3
#15 globally
knowledge
47.9
#157 globally
agentic
75.5
#4 globally
general
80.3
#2 globally
speed
58.8
#13 globally
Benchmark Scores
Compare All
Tested on 27 benchmarks · Ranked across 7 categories
Score Distribution (all 327 models)
0255075100
▲ You are here
codingCompare coding →
Mirrorcode
97.8·
Frontierswe
61.9·
Frontiercode
52.1·
reasoningCompare reasoning →
ARC-AGI

Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.

96.5·
ARC-AGI-2

ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.

83.9·
SimpleBench

Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.

71.1·
mathCompare math →
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

100.0·
FrontierMath-Tiers-1-3-v2-Private
93.4·
FrontierMath-Tier-4-v2-Private
80.5·
Quick compare:OpenAI logovs GPT-6.1 TerraAnthropic logovs Claude Fable 5.1xiaomi logovs MiMo-V2-Flash
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Anthropic Claude Sonnet
Jan 2024
30.3
n/an/actx7 benchmarks
Jan 2024
39.4+9.1
n/an/actx29 benchmarks
Jan 2024
49.2+9.8
n/an/actx27 benchmarks
Feb 2025
$3.00/M in200Kctx
May 2025
44.2-5.0
$3.00/M in200Kctx32 benchmarks
Sep 2025
43.9-0.3
$3.00/M in1.0Mctx(+800K)32 benchmarks
Sep 2025
$1.50/M in(-1.50)1.0Mctx
Feb 2026
51.3+7.4
$3.00/M in(+1.50)1.0Mctx31 benchmarks
Feb 2026
$1.50/M in(-1.50)1.0Mctx
Jun 2026
63.2+11.9
$2.00/M in(+0.50)1.0Mctx19 benchmarks
Jun 2026
$1.00/M in(-1)1.0Mctx
Sep 2026
81.4+18.2
$2.00/M in(+1)1.0Mctx27 benchmarks
Sep 2026
$1.00/M in(-1)1.0Mctx
See the full Claude Sonnet family →
Recently Happened
Claude Sonnet 5.5 added
Sep 29, 2026
Similar Models
OpenAI logo
GPT-6.1 Terra
OpenAI
81.4$1.00/1M
Anthropic logo
Claude Fable 5.1
Anthropic
81.5$10.00/1M
xiaomi logo
MiMo-V2-Flash
xiaomi
81.7$0.09/1M
Links
Info
AnthropicPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Anthropic
BenchGecko API
claude-sonnet-5-5
Specifications
  • Typemultimodal
  • Context1.0M tokens (~500 books)
  • ReleasedSep 2026
  • LicenseProprietary
  • StatusActive
  • Cost / Message~$0.014
Available On
Anthropic logoAnthropic$2.00
Categories
codingreasoningmathknowledgeagenticgeneralspeed
Learn More
context-windowtransformertokens
Share & Export
Tweet
Related Models
OpenAI logoGPT-6.1 Terra
81.4
Anthropic logoClaude Fable 5.1
81.5
xiaomi logoMiMo-V2-Flash
81.7
PaLM 2-L
81.8
Claude Sonnet 5.5 is a proprietary multimodal AI model by Anthropic, released in September 2026. It has an average benchmark score of 81.4. Context window: 1M tokens.

Related Models

GPT-6.1 Terra · OpenAIClaude Fable 5.1 · AnthropicMiMo-V2-Flash · xiaomiPaLM 2-L · UnknownClaude Opus 5 · Anthropic

Benchmarks

OTIS Mock AIME 2024-2025ProofbenchGPQA diamondMirrorcodeARC-AGI

Related Pages

Anthropic · ProviderAnthropic · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-11-14

  • Claude Sonnet 5.5 by Anthropic. BenchGecko score 81.4, rank 20 of 327 scored models (normalized average of public benchmark scores).
  • List price $2.00 input · $10.00 output per 1M tokens (as of 2026-11-14).
  • Sold by 10 providers (as of 2026-11-14): Amazon Bedrock $2.00 in / $10.00 out · Anthropic $2.00 in / $10.00 out · Azure $2.00 in / $10.00 out · Claude Platform on AWS $2.00 in / $10.00 out · Google $2.00 in / $10.00 out · and 5 more. Every provider
  • Gecko Tests · Gecko Score 69 (rank 6): Who Are You B (Knows who made it) · World Map A (97.5% of the map right) · Censorship Index A (Answers almost everything) · Knowledge Horizon A (Knows news up to May 2026) · Tokenizer Tax E (73% more tokens outside English).

How to cite · data as of 2026-11-14

Claude Sonnet 5.5 · benchmarks, pricing and providers. BenchGecko, data as of 2026-11-14. https://benchgecko.ai/model/claude-sonnet-5-5

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP