Home/Models/Claude Sonnet 5.5
C

Claude Sonnet 5.5

Closed API Anthropic 2026-09 1M context

Half the price of Opus and only 10% cheaper per task — the best-value general-purpose flagship

Intelligence v5
54.2
Leaderboard rank
#5
Input / output
$2.00/$10.00
Blended 3:1
$4.00
Output speed
135 tok/s
Value
14 pts/$
Capability profile Normalised to the field maximum
Specifications
Vendor
Anthropic
Released
2026-09
Weights
Closed, API only
Context
1,000,000 tokens
Max output
128,000 tokens
Parameters
Undisclosed
Input price
$2.00 / 1M
Output price
$10.00 / 1M
Cache hit
$0.10 / 1M
Output speed
135 tok/s
Intelligence
54.2(v5)

Benchmark scores

How it is measured →

Ranked 5 · bars are scaled to the field maximum for that benchmark.

BenchmarksScoreVs. best in fieldGap
AA Intelligence Index v5 Intelligence index 54.291% −5.2 pts
AA Intelligence Index v4.3.2 Intelligence index (legacy) 56.097% −1.6 pts
GPQA Diamond Graduate-level science reasoning 95.699% −0.8 pts
Terminal-Bench 4.0 Terminal-agent tasks 64.098% −1.2 pts
LiveBench Design Document generation (DOCX/PPT/PDF) 85.098% −1.9 pts
Vals Index Vals composite index 73.1100%

Assessment

Opus 5.5 costs twice as much per million tokens, yet only about 10% more per completed task because it writes fewer tokens. Sonnet 5.5 beats Opus on three agent-style tests: Terminal-Bench 4.0 64/60, AutomationBench-AA 72/70, AA-Briefcase 1823/1807 Elo. Available on the free claude.ai tier.

At the same price

The four models closest on blended price, side by side.

Modelv5Blendedtok/sScore gap
GPT-6.1 SolOpenAI 54.5 $4.00 77 +0.3
Gemini 4 CarbonGoogle 59.4 $4.00 87 +5.2
Gemini 4 ArgonGoogle 50.4 $4.00 80 -3.8
Muse WonderMeta 43.3 $4.00 143 -10.9

Compare side by side

Line it up against other models on price, speed and every benchmark.

The comparison tool lines up to 4 models side by side with switchable normalisation — built for purchasing decisions.

Open the comparison tool → Back to leaderboard

Related benchmarks