ai.canshu.com
中文
VRAM Calculator Advisor & TCO API Pricing Compare Models Leaderboard GPUs

Search models, GPUs, and tools

↑↓ Navigate · Enter Open · ★ Watchlist prioritized ai.canshu.com
Anthropic
Proprietary API Anthropic · claude-sonnet-5-5 · Text+Image+PDF

Claude Sonnet 5.5 Official Specifications, Pricing & Benchmarks

Most popular daily coding & agent model worldwide ($2/$10) with 1M context and 128K output.

📋 Claude Sonnet 5.5 Official API & System Card Specifications

Verified against official Anthropic developer documentation & pricing

API Model ID: claude-sonnet-5-5

1. Identity & Capabilities

API Spec
API Model ID
claude-sonnet-5-5
Release Date
2026-09-28
Weight License
Proprietary (Undisclosed)
Native Reasoning
✅ Supported
Function Calling & JSON Schema
✅ 原生支持
Batch API 异步批处理
✅ Supported (50% off)

2. Context, Output & Modalities

Limits
Max Context Window
1,000,000 tokens (1000K)
Max Output Tokens
128,000 tokens
Input Modalities
Text + Image + PDF
Output Modalities
Text
Direct China Access
🌐 Global / OpenRouter
Deployment Mode
Cloud API only (Proprietary)

3. Official Pricing & Cache

USD / 1M
Input / 1M Tokens
$ 2
Cached Input / 1M
$ 0.1 -95%
Output / 1M Tokens
$ 10
Tiered Pricing Notes
1M context window; $0.10/1M cache read; default daily driver across AI IDEs

🏆 Official Benchmark Results

Sourced from official System Card / Model Card & technical reports

Arena Elo ~1516 AA Index 55
💻 Coding
SWE-bench Verified 81.5 %

A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.

🧠 Reasoning
GPQA Diamond 89.8 %

PhD-level Google-proof science reasoning questions

🤖 Agentic
TAU2-Bench 89 %

Dual-control conversational tool-use benchmark

📚 Knowledge
MMLU-Pro 88.5 %

A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks.

🎛️ Official Sampling & Reasoning Parameters

Recommended generation settings & API parameter constraints

Extended Thinking Coding Mode
temperature=1 top_p=1 max_tokens=32,768

💡 Pair with ephemeral prompt caching on repository context to cut input spend by >90%.

📊 Claude Sonnet 5.5 Full Benchmark Scorecard

41 public benchmarks · ranks among 154 models · synced 2026-11-20

⚔️ Compare head-to-head

💻 Coding 13 items

BenchCAD (with Python tool) #1/8 96.3%

BenchCAD variant evaluated with access to a Python tool for programmatic CAD reasoning.

SWE-bench Multilingual #3/33 90.3%

A multilingual benchmark for issue resolving in software engineering that covers Java, TypeScript, JavaScript, Go, Rust, C, and C++.

SWE-Bench Pro #3/52 81.3%

SWE-Bench Pro is an advanced version of SWE-Bench that evaluates language models on complex, real-world software engineering tasks requiring extended reasoning and multi-step problem solving.

Program Bench #4/18 79.7%

Program Bench evaluates code-generation agents by asking them to recreate a program's behavior from only a compiled binary and documentation.

BenchCAD #1/8 74.7%

BenchCAD is a benchmark for programmatic CAD reasoning built from 17,900 execution-verified CadQuery programs spanning 106 industrial part families, roughly half anchored to real ISO, DIN, EN, ASME, and IEC specification tables.

DeepSWE 1.1 #15/49 71%

DeepSWE 1.1 evaluates software engineering agents using the mini-swe-agent harness.

Terminal-Bench 4.0 #1/31 70.6%

Terminal-Bench 4.0 evaluates AI agents on real-world work in terminal and command-line containerized environments across 66 tasks, with emphasis on science-adjacent and frontier engineering problems.

FrontierCode 1.1 (Extended) #1/3 64.4%

FrontierCode 1.1 Extended subset: agentic coding tasks graded for mergeable patches, reported separately from the Main subset.

FrontierSWE V2 #3/7 61.9%

FrontierSWE V2 evaluates agents on open-ended technical projects and reports mean@5. Kept separate from FrontierSWE V1 dominance-score results, which are not comparable.

Terminal-Bench-Science 0.1 #7/16 59.9%

Terminal-Bench-Science 0.1 is a Stanford-led community benchmark of 70 tasks drawn from scientific research workflows across the life, physical, earth, mathematical, and engineering sciences.

CursorBench 4.0 #3/4 55.5%

CursorBench 4.0 evaluates coding agents on long-horizon coding tasks from real Cursor sessions. Version 4.0 adds longer-horizon tasks; scores are not comparable with CursorBench 3.2.

SWE-Bench Multimodal #3/7 54.3%

SWE-Bench Multimodal extends SWE-Bench to evaluate language models on software engineering tasks that involve visual inputs such as screenshots, UI mockups, and diagrams alongside code understanding.

FrontierCode 1.1 #5/23 52.1%

FrontierCode 1.1 evaluates whether coding-agent changes are mergeable, using unit tests, maintainer-defined rubrics, and verifiers. Runs flagged for unfair internet use receive a zero score.

🤖 Agents & Tools 12 items

OSWorld 2.1 (partial) #3/7 80.1%

Partial-credit computer-use score on OSWorld 2.1, as reported in the Claude Sonnet 5.5 release. Separate from OSWorld 2.0 and strict completion scores.

Toolathlon-Verified #1/5 77.8%

Verified Toolathlon evaluation of agent tool-use capability, as reported in the MiMo-V2.6 release.

OfficeQA #3/4 76.9%

OfficeQA is a Databricks public benchmark that evaluates end-to-end grounded reasoning over a large corpus of historical U.S. Treasury Bulletin documents. Models must locate relevant tables across the corpus and perform precise numerical reasoning over them.

LatchBio SpatialBench Verified #1/3 72.5%

LatchBio SpatialBench Verified tests analysis of spatial transcriptomics data across externally validated problems, requiring the model to answer biological questions about the sample from analysis results.

OfficeQA Pro #8/15 65.6%

OfficeQA Pro evaluates AI models on professional knowledge-work questions and tasks drawn from real office workflows, including document analysis, spreadsheet reasoning, and information synthesis across business domains.

PhysicianBench #2/3 63.2%

PhysicianBench is a public benchmark of 100 physician tasks carried out in an electronic health record (EHR), measuring agentic clinical task completion.

GDPval-AA 2.1 #3/7 1844

Version 2.1 of the Artificial Analysis GDPval evaluation of professional knowledge work, reported as an Elo rating.

AA-Briefcase v1.1 #2/7 1811

AA-Briefcase v1.1 evaluates multi-hour professional office work and reports Elo scores. Kept separate from unversioned AA-Briefcase results.

LatchBio SingleCellBench #2/3 59.1%

LatchBio SingleCellBench tests analysis of single-cell RNA sequencing data across problems spanning standard workflows such as labeling cell types, finding differentially expressed genes, and correcting batch effects.

AutomationBench #11/26 44.7%

AutomationBench is a tool-use benchmark that evaluates AI agents on automating real-world workflows, testing their ability to orchestrate tools and complete multi-step automation tasks.

OSWorld 2.1 (strict) #2/4 43.5%

Strict pass rate on OSWorld 2.1: fraction of tasks where every graded checkpoint is satisfied. Separate from the OSWorld 2.1 partial-credit score and from OSWorld 2.0.

Legal Agent Benchmark #2/17 11.7%

The Legal Agent Benchmark (LAB) is Harvey's open-source benchmark for evaluating AI agents on complex, long-horizon legal work.

Source: ZeroEval / llm-stats.com (✓ = third-party verified, others vendor-reported).

Closest Self-Hostable Open-Weight Alternatives to Claude Sonnet 5.5

Open-weight models with full HuggingFace weights for private GPU / Mac deployment

Compare with these →

Frequently Asked Questions

Is Claude Sonnet 5.5 open-source or proprietary?

Claude Sonnet 5.5 (claude-sonnet-5-5) is a proprietary commercial model by Anthropic. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.

What are the context limit and API pricing of Claude Sonnet 5.5?

Claude Sonnet 5.5 supports 1000K input context and 128,000 max output tokens, priced at $2/1M input ($0.1/1M cached) and $10/1M output.