Model comparison
DeepSeek LLM 67B vs Phi-2
DeepSeek LLM 67B has enough public results to be ranked (#341); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 1 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 1 benchmark with published results for both.
Side by side
| DeepSeek LLM 67B | Phi-2 | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 24.8 | — |
| Released | 2023-11-29 | 2023-12-12 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
DeepSeek LLM 67B: 31.3 (#273), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| LMArena Coding | 1096 | — |
| HumanEval+ | — | 45.1% |
| MBPP+ | — | 54.2% |
Reasoning Not comparable
DeepSeek LLM 67B: 15.9 (#300), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| Epoch Capabilities Index | 110.54 | 107.62 |
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | 1070 | — |
| Adversarial NLI | — | 42.5% |
| BIG-Bench Hard | — | 59.4% |
| HellaSwag | — | 53.6% |
| WinoGrande | — | 54.7% |
Math Not comparable
DeepSeek LLM 67B: 8.1 (#318), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| LMArena Math | 1108 | — |
| MATH Level 5 | 6.4% | — |
Knowledge Not comparable
DeepSeek LLM 67B: 7.1 (#308), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| ARC (AI2) Challenge | — | 75.9% |
| MMLU | — | 58.4% |
| OpenBookQA | — | 73.6% |
| TriviaQA | — | 45.2% |
Multilingual Not comparable
DeepSeek LLM 67B: 29.5 (#262), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| LMArena Non-English | 1073 | — |
| LMArena Chinese | 1132 | — |
Instruction Following Not comparable
DeepSeek LLM 67B: 55.7 (#272), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| LMArena Instruction Following | 1079 | — |
Long Context Not comparable
DeepSeek LLM 67B: 33.1 (#260), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| LMArena Longer Query | 1092 | — |
Writing & Preference Not comparable
DeepSeek LLM 67B: 31.8 (#277), Phi-2: —
| Benchmark | DeepSeek LLM 67B | Phi-2 |
|---|---|---|
| LMArena Text | 1105 | — |
| LMArena Creative Writing | 1067 | — |
| LMArena Multi-Turn | 1082 | — |
Frequently asked questions
Is DeepSeek LLM 67B better than Phi-2?
DeepSeek LLM 67B has enough public results to be ranked (#341); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do DeepSeek LLM 67B and Phi-2 share?
1 benchmark has published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Phi-2 has 11.