Model comparison
Claude Sonnet 4.6 vs GPT-4
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 29.1 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 8 categories and GPT-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 4.6 leads 52.9 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 85.8% for Claude Sonnet 4.6 and 1.1% for GPT-4.
- Claude Sonnet 4.6 is cheaper at $3 / $15 per million input/output tokens, against $30 / $60 for GPT-4.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 8K.
Side by side
| Claude Sonnet 4.6 | GPT-4 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 29.1 |
| Released | 2026-02-17 | 2023-03-14 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 8K |
| Max output | 128K | 8K |
| Input $ / M tokens | $3 | $30 |
| Output $ / M tokens | $15 | $60 |
| Results tracked | 57 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.3 (#67), GPT-4: 31.6 (#283)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| WeirdML | 66.1% | 12.4% |
| LMArena Coding | 1504 | 1254 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| LMArena WebDev | 1522 | — |
| SciCode | 46.8% | — |
| BigCodeBench Instruct | — | 46% |
| BigCodeBench Complete | — | 57.2% |
| ALE-Bench | 1,327 | — |
| HumanEval+ | — | 79.3% |
Agentic & Tool Use Not comparable
Claude Sonnet 4.6: 39.1 (#28), GPT-4: —
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
| METR Time Horizons | — | 36.1% |
| Vending-Bench 2 | 7,204 | — |
Reasoning Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.1 (#45), GPT-4: 17.8 (#289)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| Chess Puzzles | 13% | 4% |
| LMArena Hard Prompts | 1484 | 1241 |
| Mystery Game Puzzles | 16% | 12% |
| DTBench | 89.9% | 62.7% |
| LMCA | 46.5% | 17.1% |
| Epoch Capabilities Index | 152.24 | 125.89 |
| ForecastBench | 62 | 57.8 |
| ARC-AGI-2 | 60.4% | — |
| NYT Connections (extended) | 80.9% | — |
| ARC-AGI-1 | 86.5% | — |
| CritPt | 3.1% | — |
| Thematic Generalization | 76.3% | — |
| BIG-Bench Hard | — | 75.1% |
| HellaSwag | — | 95.3% |
| WinoGrande | — | 87.5% |
Math Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 52.9 (#49), GPT-4: 10.8 (#309)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 1.1% |
| LMArena Math | 1462 | 1269 |
| ProofBench | 45% | — |
| MATH Level 5 | — | 23% |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
| GSM8K | — | 92% |
Knowledge Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 51.7 (#65), GPT-4: 18.4 (#282)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| GPQA Diamond | 87.4% | 35.7% |
| LMArena Expert | 1500 | 1211 |
| SimpleQA Verified | 35.5% | — |
| Vectara Hallucination Rate | 10.6% | — |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |
Multimodal Not comparable
Claude Sonnet 4.6: 38.0 (#68), GPT-4: —
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| LMArena Vision | 1283 | — |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |
Multilingual Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 54.4 (#41), GPT-4: 40.6 (#215)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| LMArena Non-English | 1440 | 1246 |
| LMArena Chinese | 1491 | 1242 |
| LMArena French | 1465 | 1283 |
| LMArena German | 1428 | 1251 |
| LMArena Japanese | 1420 | 1209 |
| LMArena Korean | 1411 | 1184 |
| LMArena Russian | 1440 | 1251 |
| LMArena Spanish | 1464 | 1261 |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.4 (#25), GPT-4: 65.3 (#222)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1475 | 1241 |
Long Context Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 45.3 (#44), GPT-4: 37.7 (#212)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1479 | 1244 |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.2 (#22), GPT-4: 34.9 (#268)
| Benchmark | Claude Sonnet 4.6 | GPT-4 |
|---|---|---|
| LMArena Text | 1458 | 1263 |
| LMArena Creative Writing | 1435 | 1244 |
| EQ-Bench Creative Writing | 1810 | 752 |
| LMArena Multi-Turn | 1464 | 1257 |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than GPT-4?
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 29.1 on the Noometry Index.
Which is cheaper, Claude Sonnet 4.6 or GPT-4?
Claude Sonnet 4.6 is cheaper. It lists at $3 per million input tokens and $15 per million output tokens; GPT-4 lists at $30 and $60.
Is Claude Sonnet 4.6 or GPT-4 better for coding?
Claude Sonnet 4.6 scores higher on coding benchmarks: 46.3 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
Claude Sonnet 4.6 does, with 1M tokens against 8K.
How many benchmarks do Claude Sonnet 4.6 and GPT-4 share?
27 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and GPT-4 has 38.