See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
Claude Sonnet 4.6
Anthropic
Each axis is a category average, normalized to a 100-point radar.
Relative edge: none clear / Relative gap: 数学推理 -11.5
Relative edge: 数学推理 +11.5 / Relative gap: none clear
Method: for each model and benchmark, the chart first averages all scores in the current mode scope instead of taking the best score, then averages those benchmark scores within each category. Only benchmarks with at least two selected models scored are included; missing values are not counted as zero.
Best overall
Claude Opus 4.6 · 211.74
Best single
Claude Opus 4.6 · GDPval-AA 1606.00
Modality coverage
Claude Sonnet 4.6 · 2 modalities
Head to head
11
Benchmarks
1
Wins
10
Losses
-144.98
Average diff
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Complete scores for each model/mode across selected benchmarks.
11 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Claude Sonnet 4.6 | Claude Opus 4.6 |
|---|---|---|
ARC-AGI-2 综合评估 | 58.30Thinking Enabled | 66.30Extended Thinking |
GPQA Diamond 综合评估 | 89.90Thinking Enabled | 91.31Extended Thinking |
HLE 综合评估 | 49.00Thinking Enabled ｜ Tools | 53.00Extended Thinking ｜ Tools |
SWE-bench Verified 编程与软件工程 | 79.60Thinking Enabled | 80.84Extended Thinking ｜ Tools |
8.3016K | 22.90Thinking Level · High | |
τ²-Bench - Telecom Agent能力评测 | 97.90Thinking Enabled ｜ Tools | 99.25Extended Thinking ｜ Tools |
BrowseComp AI Agent - 信息收集 | 74.70Thinking Enabled ｜ Tools | 84.00Thinking Enabled ｜ Tools |
OSWorld-Verified AI Agent - 工具使用 | 72.50Thinking Enabled ｜ Tools | 72.70Extended Thinking ｜ Tools |
Terminal Bench 2.0 AI Agent - 工具使用 | 59.10Thinking Enabled ｜ Tools | 65.40Extended Thinking ｜ Tools |
GDPval-AA 生产力知识 | 57.00Thinking Enabled | 1606.00Extended Thinking ｜ Tools |
Pinch Bench OpenClaw智能体能力综合测评 | 88.00Thinking Enabled ｜ Tools | 87.40Thinking Enabled ｜ Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | Claude Sonnet 4.6Anthropic | Claude Opus 4.6Anthropic |
|---|---|---|
Core specsRelease | 2026-02-17 | 2026-02-05 |
Context length | 1M | 1000K |
Max output | 8192 | 65536 |
MoE | No | No |
Supported modes | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode） | No mode data |
LicenseCode Open Source | Not provided | Not provided |
Weights Open Source | Not provided | Not provided |
Commercial use | 不开源 | 不开源 |
Modality supportText Input/Output | / | / |
Image Input/Output | / | / |
ResourcesPaper / report | Introducing Claude Sonnet 4.6 | Introducing Claude Opus 4.6 |
Claude Opus 4.6
Anthropic
