Model comparison
Claude Sonnet 4 vs GPT-5.5 Instant
GPT-5.5 Instant is the stronger model overall, scoring 42.7 to 40.8 on the Noometry Index.
Last verified . 23 shared benchmarks.
Summary
- They share 23 benchmarks with published results for both. Claude Sonnet 4 scores higher in 1 category and GPT-5.5 Instant in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 4 leads 43.3 to 26.5.
- The biggest single-benchmark swing is SciCode: 40% for Claude Sonnet 4 and 48.6% for GPT-5.5 Instant.
Side by side
| Claude Sonnet 4 | GPT-5.5 Instant | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 40.8 | 42.7 |
| Released | 2025-05-22 | 2026-05-05 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | — |
| Max output | 64K | — |
| Input $ / M tokens | $3 | — |
| Output $ / M tokens | $15 | — |
| Results tracked | 58 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Sonnet 4: 43.5 (#88), GPT-5.5 Instant: 44.3 (#74)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| SciCode | 40% | 48.6% |
| LMArena Coding | 1414 | 1433 |
| SWE-bench Verified (bash only) | 64.9% | — |
| Aider Polyglot | 61.3% | — |
| GSO | 4.9% | — |
| WeirdML | 46.1% | — |
| ALE-Bench | 655.35 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 4: 38.5 (#31), GPT-5.5 Instant: —
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| TheAgentCompany | 33.1% | — |
| Cybench | 35% | — |
| DeepResearch Bench | 46.6% | — |
| OSWorld | 43.9% | — |
| METR Time Horizons | 62% | — |
Reasoning GPT-5.5 Instant leads
Claude Sonnet 4: 22.9 (#187), GPT-5.5 Instant: 24.9 (#155)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| CritPt | 0.3% | 0% |
| LMArena Hard Prompts | 1372 | 1426 |
| Epoch Capabilities Index | 141.69 | 142.52 |
| ARC-AGI-2 | 5.9% | — |
| SimpleBench | 45.5% | — |
| Kagi LLM Benchmark | 73% | — |
| ARC-AGI-1 | 40% | — |
| Chess Puzzles | — | 12% |
| EnigmaEval | 3.1% | — |
| DTBench | 77.1% | — |
| LMCA | 29% | — |
| ForecastBench | 60.2 | — |
Math Claude Sonnet 4 leads
Claude Sonnet 4: 43.3 (#80), GPT-5.5 Instant: 26.5 (#259)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 71.1% | 68.1% |
| LMArena Math | 1375 | 1420 |
| FrontierMath (Tiers 1-3) | — | 26.3% |
| FrontierMath Tier 4 | — | 2.4% |
| Omni-MATH | 60.2% | — |
| MATH Level 5 | 84.4% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-5.5 Instant leads
Claude Sonnet 4: 41.8 (#108), GPT-5.5 Instant: 48.9 (#74)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| GPQA Diamond | 79.2% | 82.5% |
| LMArena Expert | 1372 | 1409 |
| Humanity's Last Exam | 7.8% | — |
| MMLU-Pro | 84.3% | — |
| Confabulations | 13.2% | — |
| Vectara Hallucination Rate | 10.3% | — |
| GPQA (HELM) | 70.6% | — |
Multimodal GPT-5.5 Instant leads
Claude Sonnet 4: 26.2 (#121), GPT-5.5 Instant: 40.0 (#52)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| LMArena Vision | 1191 | 1250 |
| GeoBench | 37% | — |
| VPCT | 34% | — |
| LMArena Document | — | 1403 |
| MindCube | 44.8% | — |
Multilingual GPT-5.5 Instant leads
Claude Sonnet 4: 46.7 (#156), GPT-5.5 Instant: 52.8 (#80)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| LMArena Non-English | 1333 | 1417 |
| LMArena Chinese | 1350 | 1456 |
| LMArena French | 1363 | 1428 |
| LMArena German | 1331 | 1411 |
| LMArena Japanese | 1302 | 1408 |
| LMArena Korean | 1291 | 1392 |
| LMArena Russian | 1355 | 1431 |
| LMArena Spanish | 1357 | 1429 |
Instruction Following GPT-5.5 Instant leads
Claude Sonnet 4: 71.7 (#145), GPT-5.5 Instant: 74.2 (#100)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| LMArena Instruction Following | 1376 | 1406 |
| IFEval | 84% | — |
Long Context GPT-5.5 Instant leads
Claude Sonnet 4: 33.7 (#259), GPT-5.5 Instant: 43.4 (#96)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| LMArena Longer Query | 1398 | 1422 |
| Fiction.LiveBench | 46.9% | — |
Writing & Preference GPT-5.5 Instant leads
Claude Sonnet 4: 57.1 (#132), GPT-5.5 Instant: 61.8 (#85)
| Benchmark | Claude Sonnet 4 | GPT-5.5 Instant |
|---|---|---|
| LMArena Text | 1351 | 1419 |
| LMArena Creative Writing | 1345 | 1419 |
| LMArena Multi-Turn | 1376 | 1433 |
| Short-Story Creative Writing | 81.4% | — |
| EQ-Bench Creative Writing | 1483 | — |
| WildBench | 83.8% | — |
Frequently asked questions
Is Claude Sonnet 4 better than GPT-5.5 Instant?
GPT-5.5 Instant is the stronger model overall, scoring 42.7 to 40.8 on the Noometry Index.
Is Claude Sonnet 4 or GPT-5.5 Instant better for coding?
They score almost the same on coding (43.5 vs 44.3); test both on your own repository before choosing.
How many benchmarks do Claude Sonnet 4 and GPT-5.5 Instant share?
23 benchmarks have published results for both models. Claude Sonnet 4 has 58 scored results on Noometry and GPT-5.5 Instant has 27.