Model comparison
Claude 3.7 Sonnet vs GPT-5.2 Codex
GPT-5.2 Codex is the stronger model overall, scoring 42.6 to 39.5 on the Noometry Index.
Last verified . 1 shared benchmarks.
Summary
- They share 1 benchmark with published results for both. Claude 3.7 Sonnet scores higher in 0 categories and GPT-5.2 Codex in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.2 Codex leads 41.0 to 34.1.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 52.8% for Claude 3.7 Sonnet and 72.8% for GPT-5.2 Codex.
Side by side
| Claude 3.7 Sonnet | GPT-5.2 Codex | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.5 | 42.6 |
| Released | 2025-02-24 | 2025-12-18 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.75 |
| Output $ / M tokens | — | $14 |
| Results tracked | 58 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 Codex leads
Claude 3.7 Sonnet: 40.6 (#136), GPT-5.2 Codex: 45.5 (#71)
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| SWE-bench Verified (bash only) | 52.8% | 72.8% |
| SWE-bench Verified | 61% | — |
| Aider Polyglot | 64.9% | — |
| LMArena WebDev | — | 1339 |
| SWE-bench Multilingual | — | 66.3% |
| GSO | 3.8% | — |
| LiveBench Coding | 74.5% | — |
| LMArena Coding | 1361 | — |
| CadEval | 54% | — |
| ALE-Bench | — | 1,300 |
Agentic & Tool Use GPT-5.2 Codex leads
Claude 3.7 Sonnet: 34.1 (#50), GPT-5.2 Codex: 41.0 (#22)
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| Terminal-Bench | — | 66.5% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |
Reasoning Not comparable
Claude 3.7 Sonnet: 18.6 (#277), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| ARC-AGI-2 | 0.9% | — |
| SimpleBench | 46.4% | — |
| ARC-AGI-1 | 28.6% | — |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| LMArena Hard Prompts | 1333 | — |
| LiveBench Data Analysis | 74% | — |
| Epoch Capabilities Index | 141.16 | — |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |
Math Not comparable
Claude 3.7 Sonnet: 37.5 (#153), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | — |
| Omni-MATH | 33% | — |
| LiveBench Math | 79% | — |
| LMArena Math | 1337 | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge Not comparable
Claude 3.7 Sonnet: 39.8 (#130), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| GPQA Diamond | 79.7% | — |
| Humanity's Last Exam | 8% | — |
| MMLU-Pro | 78.4% | — |
| Confabulations | 14.7% | — |
| GPQA (HELM) | 60.8% | — |
| LMArena Expert | 1321 | — |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.7 (#95), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual Not comparable
Claude 3.7 Sonnet: 44.1 (#179), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| LMArena Non-English | 1296 | — |
| LMArena Chinese | 1299 | — |
| LMArena French | 1303 | — |
| LMArena German | 1301 | — |
| LMArena Japanese | 1267 | — |
| LMArena Korean | 1249 | — |
| LMArena Russian | 1311 | — |
| LMArena Spanish | 1298 | — |
Instruction Following Not comparable
Claude 3.7 Sonnet: 72.9 (#125), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| LiveBench Instruction Following | 81.3% | — |
| IFEval | 83.4% | — |
| LMArena Instruction Following | 1352 | — |
Long Context Not comparable
Claude 3.7 Sonnet: 50.3 (#10), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| Fiction.LiveBench | 83.3% | — |
| LMArena Longer Query | 1373 | — |
Writing & Preference Not comparable
Claude 3.7 Sonnet: 54.4 (#150), GPT-5.2 Codex: —
| Benchmark | Claude 3.7 Sonnet | GPT-5.2 Codex |
|---|---|---|
| LMArena Text | 1314 | — |
| LMArena Creative Writing | 1332 | — |
| Short-Story Creative Writing | 81.1% | — |
| EQ-Bench Creative Writing | 1412 | — |
| WildBench | 81.4% | — |
| LMArena Multi-Turn | 1339 | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than GPT-5.2 Codex?
GPT-5.2 Codex is the stronger model overall, scoring 42.6 to 39.5 on the Noometry Index.
Is Claude 3.7 Sonnet or GPT-5.2 Codex better for coding?
GPT-5.2 Codex scores higher on coding benchmarks: 45.5 versus 40.6 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and GPT-5.2 Codex share?
1 benchmark has published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and GPT-5.2 Codex has 5.