Model comparison
Claude 3.7 Sonnet vs Phi-4
Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 31.2 on the Noometry Index.
Last verified . 30 shared benchmarks.
MicrosoftSummary
- They share 30 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 9 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude 3.7 Sonnet leads 37.5 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 57.8% for Claude 3.7 Sonnet and 13.8% for Phi-4.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | Phi-4 | |
|---|---|---|
| Provider | Anthropic | Microsoft |
| Noometry Index | 39.5 | 31.2 |
| Released | 2025-02-24 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 58 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 40.6 (#136), Phi-4: 34.4 (#239)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LiveBench Coding | 74.5% | 30.7% |
| LMArena Coding | 1361 | 1231 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| GSO | 3.8% | — |
| BigCodeBench Instruct | — | 45.5% |
| BigCodeBench Complete | — | 55.4% |
| CadEval | 54% | — |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 34.1 (#50), Phi-4: 22.8 (#128)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| BALROG | — | 11.6% |
| METR Time Horizons | 60% | — |
Reasoning Too close to call
Claude 3.7 Sonnet: 18.6 (#277), Phi-4: 17.7 (#291)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LiveBench Reasoning | 87.8% | 47.8% |
| LMArena Hard Prompts | 1333 | 1220 |
| LiveBench Data Analysis | 74% | 45.2% |
| Epoch Capabilities Index | 141.16 | 130.42 |
| LiveBench | 76.1% | 41.6% |
| ARC-AGI-2 | 0.9% | — |
| SimpleBench | 46.4% | — |
| ARC-AGI-1 | 28.6% | — |
| Chess Puzzles | — | 1% |
| EnigmaEval | 4.2% | — |
| ForecastBench | 61.8 | — |
Math Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 37.5 (#153), Phi-4: 20.8 (#285)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 13.8% |
| LiveBench Math | 79% | 42% |
| LMArena Math | 1337 | 1246 |
| MATH Level 5 | 91.2% | 64.9% |
| Omni-MATH | 33% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 39.8 (#130), Phi-4: 32.6 (#209)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| GPQA Diamond | 79.7% | 56.1% |
| Confabulations | 14.7% | 29.4% |
| LMArena Expert | 1321 | 1203 |
| Humanity's Last Exam | 8% | — |
| MMLU-Pro | 78.4% | — |
| Vectara Hallucination Rate | — | 3.7% |
| GPQA (HELM) | 60.8% | — |
| MMLU | — | 84.8% |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.7 (#95), Phi-4: —
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 44.1 (#179), Phi-4: 37.2 (#237)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LMArena Non-English | 1296 | 1197 |
| LMArena Chinese | 1299 | 1212 |
| LMArena French | 1303 | 1224 |
| LMArena German | 1301 | 1222 |
| LMArena Japanese | 1267 | 1158 |
| LMArena Korean | 1249 | 1151 |
| LMArena Russian | 1311 | 1209 |
| LMArena Spanish | 1298 | 1234 |
Instruction Following Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 72.9 (#125), Phi-4: 60.4 (#251)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | 81.3% | 58.4% |
| LMArena Instruction Following | 1352 | 1201 |
| IFEval | 83.4% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.3 (#10), Phi-4: 36.9 (#226)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1373 | 1217 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 54.4 (#150), Phi-4: 40.5 (#244)
| Benchmark | Claude 3.7 Sonnet | Phi-4 |
|---|---|---|
| LMArena Text | 1314 | 1217 |
| LMArena Creative Writing | 1332 | 1182 |
| Short-Story Creative Writing | 81.1% | 62.6% |
| LMArena Multi-Turn | 1339 | 1206 |
| LiveBench Language | 59.9% | 25.6% |
| EQ-Bench Creative Writing | 1412 | — |
| WildBench | 81.4% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than Phi-4?
Claude 3.7 Sonnet is the stronger model overall, scoring 39.5 to 31.2 on the Noometry Index.
Is Claude 3.7 Sonnet or Phi-4 better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 40.6 versus 34.4 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and Phi-4 share?
30 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and Phi-4 has 37.