Model comparison
DeepSeek LLM 67B vs Phi 3 Mini 4k Instruct
Phi 3 Mini 4k Instruct is the stronger model overall, scoring 27.9 to 24.9 on the Noometry Index.
Last verified . 11 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 11 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 6 categories and Phi 3 Mini 4k Instruct in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Mini 4k Instruct leads 28.5 to 7.0.
Side by side
| DeepSeek LLM 67B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 24.9 | 27.9 |
| Released | 2023-11-29 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 15 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.9 (#278), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1096 | 1093 |
| LiveBench Coding | — | 15.5% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Reasoning DeepSeek LLM 67B leads
DeepSeek LLM 67B: 16.5 (#304), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1070 | 1072 |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 110.5 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Phi 3 Mini 4k Instruct leads
DeepSeek LLM 67B: 8.7 (#324), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1108 | 1111 |
| OTIS Mock AIME 2024-2025 | 0.8% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 6.4% | — |
Knowledge Phi 3 Mini 4k Instruct leads
DeepSeek LLM 67B: 7.0 (#313), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| GPQA Diamond | 24.6% | — |
| LMArena Expert | — | 1045 |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual DeepSeek LLM 67B leads
DeepSeek LLM 67B: 29.4 (#267), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1073 | 1021 |
| LMArena Chinese | 1132 | 1021 |
| LMArena French | — | 1076 |
| LMArena German | — | 1044 |
| LMArena Japanese | — | 935 |
| LMArena Korean | — | 905 |
| LMArena Russian | — | 1022 |
| LMArena Spanish | — | 1085 |
Instruction Following DeepSeek LLM 67B leads
DeepSeek LLM 67B: 55.4 (#277), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1079 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context DeepSeek LLM 67B leads
DeepSeek LLM 67B: 33.1 (#265), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1092 | 1044 |
Writing & Preference DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.6 (#282), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | DeepSeek LLM 67B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1105 | 1073 |
| LMArena Creative Writing | 1067 | 1037 |
| LMArena Multi-Turn | 1082 | 1018 |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is DeepSeek LLM 67B better than Phi 3 Mini 4k Instruct?
Phi 3 Mini 4k Instruct is the stronger model overall, scoring 27.9 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Phi 3 Mini 4k Instruct better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 26.6 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Phi 3 Mini 4k Instruct share?
11 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.