Model comparison
Claude 3.5 Sonnet vs GPT-6 Astra
GPT-6 Astra is the stronger model overall, scoring 70.8 to 34.6 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 0 categories and GPT-6 Astra in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Astra leads 93.5 to 19.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 100% for GPT-6 Astra.
Side by side
| Claude 3.5 Sonnet | GPT-6 Astra | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.6 | 70.8 |
| Released | 2024-06-20 | 2026-09-03 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 128K |
| Input $ / M tokens | — | $10 |
| Output $ / M tokens | — | $50 |
| Results tracked | 60 | 56 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Astra leads
Claude 3.5 Sonnet: 39.0 (#165), GPT-6 Astra: 73.7 (#2)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| GSO | 4.6% | 79.4% |
| WeirdML | 40% | 93.6% |
| LMArena Coding | 1342 | 1487 |
| DeepSWE | — | 74.1% |
| FrontierCode | — | 53.3% |
| Aider Polyglot | 51.6% | — |
| LMArena WebDev | — | 1786 |
| FrontierSWE | — | 65.5% |
| SciCode | — | 56.5% |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| MirrorCode | — | 46.7% |
| BigCodeBench Complete | 58.6% | — |
| CadEval | 48% | — |
| ALE-Bench | — | 2,951 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use GPT-6 Astra leads
Claude 3.5 Sonnet: 32.3 (#67), GPT-6 Astra: 52.9 (#3)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| BALROG | 32.6% | 68.3% |
| APEX-Agents | — | 64.7% |
| Remote Labor Index | — | 20.8% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| GDP.pdf | — | 34.2% |
| METR Time Horizons | 45.2% | — |
| Vending-Bench 2 | — | 15,515 |
Reasoning GPT-6 Astra leads
Claude 3.5 Sonnet: 23.1 (#183), GPT-6 Astra: 85.1 (#1)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| LMArena Hard Prompts | 1305 | 1462 |
| DTBench | 67.8% | 97.3% |
| Epoch Capabilities Index | 133.55 | 166.45 |
| ARC-AGI-2 | — | 95% |
| SimpleBench | 41.4% | — |
| NYT Connections (extended) | — | 98.1% |
| ARC-AGI-1 | — | 98.5% |
| CritPt | — | 31.7% |
| Chess Puzzles | — | 72% |
| EnigmaEval | 0.9% | — |
| EBR-Bench | — | 76.2% |
| LiveBench Reasoning | 56.7% | — |
| Mystery Game Puzzles | — | 84% |
| LiveBench Data Analysis | 55% | — |
| LMCA | — | 64.4% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | 60.7 | — |
| LiveBench | 59% | — |
Math GPT-6 Astra leads
Claude 3.5 Sonnet: 19.2 (#288), GPT-6 Astra: 93.5 (#2)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 100% |
| LMArena Math | 1307 | 1465 |
| FrontierMath (Tiers 1-3) | — | 93.7% |
| FrontierMath Tier 4 | — | 97.6% |
| ProofBench | — | 99% |
| Omni-MATH | 27.6% | — |
| LiveBench Math | 52.3% | — |
| MATH Level 5 | 56.9% | — |
| FrontierMath (Feb 2025 set) | 2.1% | — |
| FrontierMath Erdős | — | 2.9% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-6 Astra leads
Claude 3.5 Sonnet: 28.6 (#245), GPT-6 Astra: 75.3 (#1)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| GPQA Diamond | 55.3% | 95.8% |
| Humanity's Last Exam | 4.1% | 54.8% |
| LMArena Expert | 1265 | 1483 |
| SimpleQA Verified | — | 75.6% |
| MMLU-Pro | 77.7% | — |
| Confabulations | 19.9% | — |
| Vectara Hallucination Rate | — | 8.7% |
| GPQA (HELM) | 56.5% | — |
| MMLU | 87.3% | — |
Multimodal GPT-6 Astra leads
Claude 3.5 Sonnet: 26.5 (#120), GPT-6 Astra: 55.0 (#3)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| LMArena Vision | 1125 | 1281 |
| Video-MME | 60% | — |
| GeoBench | 62% | — |
| VPCT | 33% | — |
| Blueprint-Bench 2 | — | 49.7% |
| Furniture Assembly | — | 80% |
| LMArena Document | — | 1468 |
Multilingual GPT-6 Astra leads
Claude 3.5 Sonnet: 43.2 (#185), GPT-6 Astra: 53.7 (#61)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| LMArena Non-English | 1283 | 1430 |
| LMArena Chinese | 1272 | 1484 |
| LMArena French | 1305 | 1456 |
| LMArena German | 1297 | 1440 |
| LMArena Japanese | 1234 | 1379 |
| LMArena Korean | 1200 | 1426 |
| LMArena Russian | 1306 | 1436 |
| LMArena Spanish | 1290 | 1407 |
Instruction Following GPT-6 Astra leads
Claude 3.5 Sonnet: 68.8 (#182), GPT-6 Astra: 76.3 (#44)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| LMArena Instruction Following | 1297 | 1450 |
| LiveBench Instruction Following | 69.3% | — |
| IFEval | 85.5% | — |
Long Context GPT-6 Astra leads
Claude 3.5 Sonnet: 39.9 (#167), GPT-6 Astra: 44.5 (#62)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| LMArena Longer Query | 1311 | 1456 |
Writing & Preference GPT-6 Astra leads
Claude 3.5 Sonnet: 52.9 (#164), GPT-6 Astra: 75.3 (#7)
| Benchmark | Claude 3.5 Sonnet | GPT-6 Astra |
|---|---|---|
| LMArena Text | 1298 | 1441 |
| LMArena Creative Writing | 1292 | 1418 |
| EQ-Bench Creative Writing | 1451 | 2173 |
| LMArena Multi-Turn | 1326 | 1448 |
| Short-Story Creative Writing | 80.3% | — |
| WildBench | 79.2% | — |
| LiveBench Language | 53.8% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than GPT-6 Astra?
GPT-6 Astra is the stronger model overall, scoring 70.8 to 34.6 on the Noometry Index.
Is Claude 3.5 Sonnet or GPT-6 Astra better for coding?
GPT-6 Astra scores higher on coding benchmarks: 73.7 versus 39.0 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and GPT-6 Astra share?
27 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and GPT-6 Astra has 56.