メインコンテンツへスキップ

Claude Sonnet 5.5 (Max, Default Fallback)

AnthropicClaudeProprietary

説明

Claude Sonnet 5.5 is the second model in Anthropic's Claude 5.5 family and a faster, lower-cost complement to Claude Opus 5.5. It targets well-scoped everyday tasks, coding, and polished documents, slides, and spreadsheets, and runs 30%+ faster than Claude Sonnet 5 while typically using far fewer tokens at the same $2/$10 per million input/output prices. It accepts text and image input with text output, supports multilingual and vision workloads, tool use, and adaptive thinking (Claude Platform default effort high; Claude Code and apps default medium). Context window is 1M tokens with 128K max output on the sync Messages API; reliable knowledge and training-data cutoffs are June 2026. Cache reads are $0.10 and 5-minute cache writes are $2.50 per million tokens. Capability scores are self-reported from the Claude Sonnet 5.5 System Card (Table 8.1.A and §8) and the launch announcement. Available on the Claude API as `claude-sonnet-5-5`.

リリース日
2026-09-28
パラメータ
—
コンテキスト長
1.0M
モダリティ
image, pdf, text

能力レーダー

55
general
70
coding
70
reasoning
51
science推定
60
agents
80
multimodal

専用の科学ベンチマークがない場合、Science は LLM Stats の科学スコアまたは推論能力から推定します。

ランキング

ベンチマークスコア (LLM Stats)

(LLM Stats (zeroeval))

Agents

AA-Briefcase v1.11811.00 / 3000自己申告
Program Bench79.7%自己申告
Toolathlon-Verified77.8%自己申告
OfficeQA76.9%自己申告
OfficeQA Pro65.6%自己申告
AutomationBench44.7%自己申告

Biology

BioMysteryBench89.2%自己申告
De novo protein binder design82.3%自己申告
LatchBio SpatialBench Verified72.5%自己申告
Biomedical image analysis72.2%自己申告
Protocols Troubleshooting67.3%自己申告
Protocols Understanding (V2)66.6%自己申告
Medicinal Chemistry (ADME)65.3%自己申告
Protein Design — Library Ranking54.8%自己申告
Protein Design — Sequence Generation51.0%自己申告
BioMysteryBench (Human Difficult)44.7%自己申告
Morphology-to-molecule matching25.0%自己申告

Code

BenchCAD (with Python tool)96.3%自己申告
BenchCAD74.7%自己申告
DeepSWE 1.171.0%自己申告
FrontierSWE V261.9%自己申告
CursorBench 4.055.5%自己申告

General

GDPval-AA 2.11844.00 / 3000自己申告
Global-MMLU92.1%自己申告

Healthcare

HealthBench Professional (raw)77.1%自己申告
HealthBench (raw)69.4%自己申告
HealthBench Professional69.2%自己申告
HealthBench65.4%自己申告
PhysicianBench63.2%自己申告

Language

MILU91.6%自己申告

Legal

Legal Agent Benchmark11.7%自己申告

Math

ArXivMath (with tools)95.2%自己申告
ArXivMath86.8%自己申告

Multimodal

OSWorld 2.1 (partial)80.1%自己申告
OSWorld 2.1 (strict)43.5%自己申告

Reasoning

SWE-bench Multilingual90.3%自己申告
SWE-Bench ProPrinceton NLP (2024)81.3%自己申告
Terminal-Bench 4.070.6%自己申告
Humanity's Last Exam (with tools)64.5%自己申告
FrontierCode 1.1 (Extended)64.4%自己申告
Terminal-Bench-Science 0.159.9%自己申告
Humanity's Last Exam (no tools, text-only)56.9%自己申告
SWE-Bench Multimodal54.3%自己申告
FrontierCode 1.152.1%自己申告

Science

LatchBio SingleCellBench59.1%自己申告

Vision

Chartography90.2%自己申告
Chartography (no tools)61.6%自己申告

AA評価指数

(Artificial Analysis)
Engineering(Artificial Analysis)
58.5
Strategy And Ops
58.1
Finance And Accounting
55.8
Intelligence Index(Artificial Analysis)
54.5
Healthcare And Medical
54.2
Economics
54.1
Legal
52.2

LLM Statsカテゴリスコア

(LLM Stats (zeroeval))
Productivity
100
Agents
100
Reasoning
100
General
90
Language
90
Math
70
Multimodal
70
Healthcare
70
Chemistry
70
Code
70
Vision
70
Science
60
Biology
60
Tool Calling
60
Legal
10

価格設定

入力価格$2 / 1Mトークン
出力価格$10 / 1Mトークン
混合価格(3:1)$4 / 1Mトークン
キャッシュ読み取り価格$0.1 / 1Mトークン
キャッシュ書き込み価格$2.5 / 1Mトークン

速度

トークン/秒132.8
初トークン遅延434.00s
初回答遅延434.00s

プロバイダー価格ランキング

プロバイダー価格ランキング

1 プロバイダー

プロバイダー入力出力
1Anthropicプライマリ
$2
$10

このモデルの異なるAPIプロバイダー間の価格を比較。

外部リンク