Model comparison
DeepSeek-V3 vs Phi-2
DeepSeek-V3 has enough public results to be ranked (#166); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 9 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 9 benchmarks with published results for both.
Side by side
| DeepSeek-V3 | Phi-2 | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 39.5 | — |
| Released | 2024-12-26 | 2023-12-12 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 164K | — |
| Input $ / M tokens | $0.24 | — |
| Output $ / M tokens | $0.90 | — |
| Results tracked | 60 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
DeepSeek-V3: 42.3 (#106), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| HumanEval+ | 86.6% | 45.1% |
| MBPP+ | 73% | 54.2% |
| Aider Polyglot | 55.1% | — |
| SciCode | 35.8% | — |
| WeirdML | 36.1% | — |
| BigCodeBench Instruct | 50% | — |
| LiveBench Coding | 70.9% | — |
| LMArena Coding | 1368 | — |
| BigCodeBench Complete | 62.2% | — |
Agentic & Tool Use Not comparable
DeepSeek-V3: —, Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| METR Time Horizons | 49.6% | — |
Reasoning Not comparable
DeepSeek-V3: 20.5 (#236), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| BIG-Bench Hard | 87.5% | 59.4% |
| Epoch Capabilities Index | 135.94 | 107.94 |
| HellaSwag | 88.9% | 53.6% |
| WinoGrande | 85.2% | 54.7% |
| SimpleBench | 27.2% | — |
| Kagi LLM Benchmark | 52.3% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | 65.8% | — |
| LMArena Hard Prompts | 1365 | — |
| DTBench | 64.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 15.5% | — |
| Adversarial NLI | — | 42.5% |
| ForecastBench | 59.1 | — |
| LiveBench | 66.9% | — |
| PIQA | 84.7% | — |
Math Not comparable
DeepSeek-V3: 32.1 (#219), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| Omni-MATH | 40.3% | — |
| LiveBench Math | 73.5% | — |
| LMArena Math | 1373 | — |
| MATH Level 5 | 75.5% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Not comparable
DeepSeek-V3: 37.5 (#155), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| ARC (AI2) Challenge | 95.3% | 75.9% |
| MMLU | 87.2% | 58.4% |
| TriviaQA | 82.9% | 45.2% |
| GPQA Diamond | 67.6% | — |
| MMLU-Pro | 72.3% | — |
| Confabulations | 26.1% | — |
| Vectara Hallucination Rate | 6.1% | — |
| GPQA (HELM) | 53.8% | — |
| LMArena Expert | 1351 | — |
| OpenBookQA | — | 73.6% |
Multilingual Not comparable
DeepSeek-V3: 48.5 (#143), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| LMArena Non-English | 1358 | — |
| LMArena Chinese | 1391 | — |
| LMArena French | 1385 | — |
| LMArena German | 1374 | — |
| LMArena Japanese | 1333 | — |
| LMArena Korean | 1319 | — |
| LMArena Russian | 1373 | — |
| LMArena Spanish | 1358 | — |
Instruction Following Not comparable
DeepSeek-V3: 72.8 (#130), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| LiveBench Instruction Following | 81.5% | — |
| IFEval | 83.2% | — |
| LMArena Instruction Following | 1345 | — |
Long Context Not comparable
DeepSeek-V3: 34.0 (#253), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| Fiction.LiveBench | 50% | — |
| LMArena Longer Query | 1352 | — |
Writing & Preference Not comparable
DeepSeek-V3: 57.4 (#130), Phi-2: —
| Benchmark | DeepSeek-V3 | Phi-2 |
|---|---|---|
| LMArena Text | 1375 | — |
| LMArena Creative Writing | 1364 | — |
| Short-Story Creative Writing | 77% | — |
| EQ-Bench Creative Writing | 1472 | — |
| WildBench | 83% | — |
| LMArena Multi-Turn | 1389 | — |
| LiveBench Language | 49.1% | — |
Frequently asked questions
Is DeepSeek-V3 better than Phi-2?
DeepSeek-V3 has enough public results to be ranked (#166); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do DeepSeek-V3 and Phi-2 share?
9 benchmarks have published results for both models. DeepSeek-V3 has 60 scored results on Noometry and Phi-2 has 11.