Model comparison
Claude Sonnet 4.6 vs o3
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 47.5 on the Noometry Index. o3 costs 1.7× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.
Last verified . 39 shared benchmarks.
Summary
- They share 39 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 6 categories and o3 in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 32.0.
- The biggest single-benchmark swing is ARC-AGI-2: 60.4% for Claude Sonnet 4.6 and 6.5% for o3.
- o3 is cheaper at $2 / $8 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 200K.
Side by side
| Claude Sonnet 4.6 | o3 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 47.5 |
| Released | 2026-02-17 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $3 | $2 |
| Output $ / M tokens | $15 | $8 |
| Results tracked | 57 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Claude Sonnet 4.6: 46.3 (#67), o3: 46.8 (#64)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| SWE-bench Verified | 75.2% | 62.3% |
| WeirdML | 66.1% | 52.4% |
| LMArena Coding | 1504 | 1408 |
| ALE-Bench | 1,327 | 933.55 |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| SWE-bench Verified (bash only) | — | 58.4% |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1522 | — |
| SciCode | 46.8% | — |
| GSO | — | 8.8% |
| CadEval | — | 74% |
Agentic & Tool Use Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 39.1 (#28), o3: 34.5 (#44)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| DeepResearch Bench | 54.9% | 45.2% |
| OSWorld | 72.1% | 23% |
| LMArena Search | 1221 | 1144 |
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| Berkeley Function Calling Leaderboard | — | 63% |
| OSWorld 2.0 | 9.3% | — |
| GDPval | — | 30.8% |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| METR Time Horizons | — | 65.4% |
| Vending-Bench 2 | 7,204 | — |
Reasoning Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 46.1 (#45), o3: 32.0 (#78)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| ARC-AGI-2 | 60.4% | 6.5% |
| ARC-AGI-1 | 86.5% | 60.8% |
| CritPt | 3.1% | 1.4% |
| Chess Puzzles | 13% | 38% |
| LMArena Hard Prompts | 1484 | 1402 |
| Mystery Game Puzzles | 16% | 29% |
| DTBench | 89.9% | 84.8% |
| LMCA | 46.5% | 39.7% |
| Epoch Capabilities Index | 152.24 | 146.86 |
| ForecastBench | 62 | 62.5 |
| SimpleBench | — | 53.1% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 80.9% | — |
| EnigmaEval | — | 13.1% |
| Thematic Generalization | 76.3% | — |
Math Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 52.9 (#49), o3: 50.2 (#58)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 84.4% |
| LMArena Math | 1462 | 1426 |
| FrontierMath (Feb 2025 set) | 32.4% | 18.7% |
| FrontierMath Tier 4 (v1) | 8.3% | 2.1% |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| ProofBench | 45% | — |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |
Knowledge o3 leads
Claude Sonnet 4.6: 51.7 (#65), o3: 54.6 (#52)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| GPQA Diamond | 87.4% | 81.8% |
| SimpleQA Verified | 35.5% | 49.4% |
| LMArena Expert | 1500 | 1402 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| Vectara Hallucination Rate | 10.6% | — |
| GPQA (HELM) | — | 75.3% |
Multimodal o3 leads
Claude Sonnet 4.6: 38.0 (#68), o3: 41.4 (#36)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| LMArena Vision | 1283 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |
Multilingual Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 54.4 (#41), o3: 51.7 (#105)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| LMArena Non-English | 1440 | 1401 |
| LMArena Chinese | 1491 | 1437 |
| LMArena French | 1465 | 1430 |
| LMArena German | 1428 | 1420 |
| LMArena Japanese | 1420 | 1403 |
| LMArena Korean | 1411 | 1370 |
| LMArena Russian | 1440 | 1406 |
| LMArena Spanish | 1464 | 1395 |
Instruction Following Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 77.4 (#25), o3: 72.8 (#127)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| LMArena Instruction Following | 1475 | 1368 |
| IFEval | — | 86.9% |
Long Context o3 leads
Claude Sonnet 4.6: 45.3 (#44), o3: 53.3 (#6)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| LMArena Longer Query | 1479 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |
Writing & Preference Claude Sonnet 4.6 leads
Claude Sonnet 4.6: 70.2 (#22), o3: 63.5 (#64)
| Benchmark | Claude Sonnet 4.6 | o3 |
|---|---|---|
| LMArena Text | 1458 | 1410 |
| LMArena Creative Writing | 1435 | 1359 |
| EQ-Bench Creative Writing | 1810 | 1676 |
| LMArena Multi-Turn | 1464 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| WildBench | — | 86.1% |
| EQ-Bench 4 | 1207 | — |
Frequently asked questions
Is Claude Sonnet 4.6 better than o3?
Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 47.5 on the Noometry Index. o3 costs 1.7× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 4.6 or o3?
o3 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.
Is Claude Sonnet 4.6 or o3 better for coding?
They score almost the same on coding (46.3 vs 46.8); test both on your own repository before choosing.
Which has the bigger context window?
Claude Sonnet 4.6 does, with 1M tokens against 200K.
How many benchmarks do Claude Sonnet 4.6 and o3 share?
39 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and o3 has 63.