Model comparison
Claude 3.5 Sonnet vs Phi 3 Mini 4k Instruct
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 27.9 on the Noometry Index.
Last verified . 27 shared benchmarks.
MicrosoftSummary
- They share 27 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 7 categories and Phi 3 Mini 4k Instruct in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude 3.5 Sonnet leads 52.9 to 27.6.
- The biggest single-benchmark swing is LiveBench Coding: 67.1% for Claude 3.5 Sonnet and 15.5% for Phi 3 Mini 4k Instruct.
- Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Anthropic | Microsoft |
| Noometry Index | 34.6 | 27.9 |
| Released | 2024-06-20 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 60 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.0 (#165), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Coding | 67.1% | 15.5% |
| LMArena Coding | 1342 | 1093 |
| HumanEval+ | 81.7% | 59.1% |
| MBPP+ | 74.3% | 54.2% |
| Aider Polyglot | 51.6% | — |
| GSO | 4.6% | — |
| WeirdML | 40% | — |
| BigCodeBench Instruct | 46.8% | — |
| BigCodeBench Complete | 58.6% | — |
| CadEval | 48% | — |
Agentic & Tool Use Not comparable
Claude 3.5 Sonnet: 32.3 (#67), Phi 3 Mini 4k Instruct: —
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| BALROG | 32.6% | — |
| METR Time Horizons | 45.2% | — |
Reasoning Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 23.1 (#183), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Reasoning | 56.7% | 26.8% |
| LMArena Hard Prompts | 1305 | 1072 |
| LiveBench Data Analysis | 55% | 34.7% |
| LiveBench | 59% | 22.4% |
| SimpleBench | 41.4% | — |
| Chess Puzzles | — | 0% |
| EnigmaEval | 0.9% | — |
| DTBench | 67.8% | — |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 133.55 | — |
| ForecastBench | 60.7 | — |
| HellaSwag | — | 76.7% |
| WinoGrande | — | 70.8% |
Math Phi 3 Mini 4k Instruct leads
Claude 3.5 Sonnet: 19.2 (#288), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Math | 52.3% | 15.7% |
| LMArena Math | 1307 | 1111 |
| OTIS Mock AIME 2024-2025 | 8.5% | — |
| Omni-MATH | 27.6% | — |
| MATH Level 5 | 56.9% | — |
| FrontierMath (Feb 2025 set) | 2.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Too close to call
Claude 3.5 Sonnet: 28.6 (#245), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1265 | 1045 |
| MMLU | 87.3% | 68.8% |
| GPQA Diamond | 55.3% | — |
| Humanity's Last Exam | 4.1% | — |
| MMLU-Pro | 77.7% | — |
| Confabulations | 19.9% | — |
| GPQA (HELM) | 56.5% | — |
| ARC (AI2) Challenge | — | 84.9% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multimodal Not comparable
Claude 3.5 Sonnet: 26.5 (#120), Phi 3 Mini 4k Instruct: —
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Vision | 1125 | — |
| Video-MME | 60% | — |
| GeoBench | 62% | — |
| VPCT | 33% | — |
Multilingual Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 43.2 (#185), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1283 | 1021 |
| LMArena Chinese | 1272 | 1021 |
| LMArena French | 1305 | 1076 |
| LMArena German | 1297 | 1044 |
| LMArena Japanese | 1234 | 935 |
| LMArena Korean | 1200 | 905 |
| LMArena Russian | 1306 | 1022 |
| LMArena Spanish | 1290 | 1085 |
Instruction Following Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 68.8 (#182), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Instruction Following | 69.3% | 39.1% |
| LMArena Instruction Following | 1297 | 1053 |
| IFEval | 85.5% | — |
Long Context Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.9 (#167), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1311 | 1044 |
Writing & Preference Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 52.9 (#164), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Claude 3.5 Sonnet | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1298 | 1073 |
| LMArena Creative Writing | 1292 | 1037 |
| LMArena Multi-Turn | 1326 | 1018 |
| LiveBench Language | 53.8% | 9.2% |
| Short-Story Creative Writing | 80.3% | — |
| EQ-Bench Creative Writing | 1451 | — |
| WildBench | 79.2% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than Phi 3 Mini 4k Instruct?
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 27.9 on the Noometry Index.
Is Claude 3.5 Sonnet or Phi 3 Mini 4k Instruct better for coding?
Claude 3.5 Sonnet scores higher on coding benchmarks: 39.0 versus 26.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and Phi 3 Mini 4k Instruct share?
27 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.