Claude Sonnet 4.6vsGPT-5.2
Claude Sonnet 4.6 and GPT-5.2 are tied across 10 shared benchmarks: Claude Sonnet 4.6 leads on 5, GPT-5.2 leads on 5, with 0 ties and an average score difference of +1.61.
Claude Sonnet 4.6
Anthropic · 2026-02-17 · Chat model
GPT-5.2
OpenAI · 2025-12-11 · Chat model
Claude Sonnet 4.65 wins(50%)(50%)5 winsGPT-5.2
Benchmark scores
Grouped by capability, sorted by largest gap within each. 10 shared benchmarks.
General Knowledge
Claude Sonnet 4.6 3/4| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| LiveBench | 75.4712 / 115Thinking Medium (No Tools) | 48.9194 / 115Normal (No Tools) | +26.56 |
| ARC-AGI-2 | 58.3021 / 62 | 54.2023 / 62深度思考（无工具、并行） | +4.10 |
| HLE | 4932 / 172 | 45.5041 / 172Deep Thinking (With Tools + Internet) | +3.50 |
| GPQA Diamond | 89.9024 / 187 | 93.209 / 187深度思考（无工具、并行） | -3.30 |
Agent Level Benchmark
GPT-5.2 1/1| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| τ²-Bench - Telecom | 97.909 / 35 | 98.704 / 35极高强度思考（工具） | -0.80 |
AI Agent - Information Search
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| BrowseComp | 74.7027 / 53 | 65.8031 / 53Deep Thinking (With Tools + Internet) | +8.90 |
AI Agent - Tool Usage
Claude Sonnet 4.6 1/1| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| MCP-Atlas | 69.5017 / 27Normal (With Tools) | 67.6018 / 27极高强度思考（工具） | +1.90 |
Coding and Software Engineer
GPT-5.2 1/1| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| SWE-bench Verified | 79.6018 / 112 | 8017 / 112极高强度思考（工具） | -0.40 |
Math and Reasoning
GPT-5.2 1/1| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| FrontierMath - Tier 4 | 8.3034 / 80Thinking (No Tools, 16K Budget) | 18.8016 / 80Thinking High (No Tools) | -10.50 |
Productivity Knowledge
GPT-5.2 1/1| Benchmark | Claude Sonnet 4.6 | GPT-5.2 | Diff |
|---|---|---|---|
| GDPval-AA | 5711 / 21 | 70.909 / 21Thinking High (With Tools) | -13.90 |
Specs
| Field | Claude Sonnet 4.6 | GPT-5.2 |
|---|---|---|
| Publisher | Anthropic | OpenAI |
| Release date | 2026-02-17 | 2025-12-11 |
| Model type | Chat model | Chat model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 400K |
| Max output | 8K | Not available |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Claude Sonnet 4.6 | GPT-5.2 |
|---|---|---|
| Text input | $3 / 1M tokens | $1.75 / 1M tokens |
| Text output | $15 / 1M tokens | $14 / 1M tokens |
| Cache read | $0.3 / 1M tokens | $0.175 / 1M tokens |
| Cache write | $3.75 / 1M tokens | $1.75 / 1M tokens |
Summary
- Claude Sonnet 4.6leads in:General Knowledge (3/4), AI Agent - Information Search (1/1), AI Agent - Tool Usage (1/1)
- GPT-5.2leads in:Agent Level Benchmark (1/1), Coding and Software Engineer (1/1), Math and Reasoning (1/1), Productivity Knowledge (1/1)
On average across the 10 shared benchmarks, Claude Sonnet 4.6 scores 1.61 higher.
Largest single-benchmark gap: LiveBench — Claude Sonnet 4.6 75.47 vs GPT-5.2 48.91 (+26.56).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.