GPT-5.1vsClaude Sonnet 4.5
Across 17 shared benchmarks, GPT-5.1 leads overall: GPT-5.1 wins 10, Claude Sonnet 4.5 wins 7, with 0 ties and an average score difference of +3.72.
GPT-5.1
OpenAI · 2025-11-12 · Reasoning model
Claude Sonnet 4.5
Anthropic · 2025-09-30 · Chat model
GPT-5.110 wins(59%)(41%)7 winsClaude Sonnet 4.5
Benchmark scores
Grouped by capability, sorted by largest gap within each. 17 shared benchmarks.
General Knowledge
GPT-5.1 3/5| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| LiveBench | 42.65106 / 115Normal (No Tools) | 53.6983 / 115Normal (No Tools) | -11.04 |
| ARC-AGI | 72.8028 / 68 | 63.7035 / 68 | +9.10 |
| HLE | 26.5097 / 172 | 33.6080 / 172 | -7.10 |
| GPQA Diamond | 88.1031 / 187 | 83.4063 / 187 | +4.70 |
| ARC-AGI-2 | 17.6036 / 62 | 13.6038 / 62 | +4 |
Math and Reasoning
GPT-5.1 2/3| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| FrontierMath | 26.7013 / 60Thinking High (With Tools) | 5.2038 / 60 | +21.50 |
| FrontierMath - Tier 4 | 12.5029 / 80Thinking High (With Tools) | 2.1056 / 80Normal (No Tools) | +10.40 |
| AIME2025 | 9428 / 107 | 1001 / 107 | -6 |
Agent Level Benchmark
Even 2/2| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| Terminal Bench Hard | 432 / 13Thinking High (With Tools) | 338 / 13 | +10 |
| τ²-Bench - Telecom | 95.6014 / 35Thinking High (With Tools) | 985 / 35 | -2.40 |
AI Agent - Tool Usage
Even 2/2| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| MCP-Atlas | 50.1025 / 27Thinking High (With Tools) | 59.5021 / 27Thinking (With Tools) | -9.40 |
| Terminal Bench 2.0 | 47.6038 / 47Thinking High (With Tools) | 42.8042 / 47 | +4.80 |
Coding and Software Engineer
Even 2/2| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| SWE-Bench Pro - Public | 50.8040 / 54Thinking High (No Tools) | 43.6047 / 54 | +7.20 |
| SWE-bench Verified | 76.3034 / 112 | 828 / 112 | -5.70 |
AI Agent - Information Search
GPT-5.1 1/1| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| BrowseComp | 50.8043 / 53Thinking High (No Tools) | 24.1051 / 53 | +26.70 |
Commonsense Reasoning
Claude Sonnet 4.5 1/1| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| Simple Bench | 53.2023 / 63Thinking High (No Tools) | 54.3022 / 63Normal (No Tools) | -1.10 |
Multimodal Understanding
GPT-5.1 1/1| Benchmark | GPT-5.1 | Claude Sonnet 4.5 | Diff |
|---|---|---|---|
| MMMU | 85.402 / 29 | 77.8015 / 29 | +7.60 |
Specs
| Field | GPT-5.1 | Claude Sonnet 4.5 |
|---|---|---|
| Publisher | OpenAI | Anthropic |
| Release date | 2025-11-12 | 2025-09-30 |
| Model type | Reasoning model | Chat model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 400K | 1000K |
| Max output | 128K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.1 | Claude Sonnet 4.5 |
|---|---|---|
| Text input | $1.25 / 1M tokens | $3 / 1M tokens |
| Text output | $10 / 1M tokens | $15 / 1M tokens |
| Cache read | $0.125 / 1M tokens | $0.3 / 1M tokens |
| Cache write | $0 / 1M tokens | $3.75 / 1M tokens |
Summary
- GPT-5.1leads in:General Knowledge (3/5), Math and Reasoning (2/3), AI Agent - Information Search (1/1), Multimodal Understanding (1/1)
- Claude Sonnet 4.5leads in:Commonsense Reasoning (1/1)
- Tied in:Agent Level Benchmark, AI Agent - Tool Usage, Coding and Software Engineer
On average across the 17 shared benchmarks, GPT-5.1 scores 3.72 higher.
Largest single-benchmark gap: BrowseComp — GPT-5.1 50.80 vs Claude Sonnet 4.5 24.10 (+26.70).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.