Model comparison
DeepSeek-R1-Distill-Llama-70B vs Phi 3 Mini 4k Instruct
DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.8 to 27.9 on the Noometry Index.
Last verified . 7 shared benchmarks.
DeepSeek
MicrosoftSummary
- They share 7 benchmarks with published results for both. DeepSeek-R1-Distill-Llama-70B scores higher in 6 categories and Phi 3 Mini 4k Instruct in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-R1-Distill-Llama-70B leads 49.0 to 27.6.
- The biggest single-benchmark swing is LiveBench Math: 58.1% for DeepSeek-R1-Distill-Llama-70B and 15.7% for Phi 3 Mini 4k Instruct.
Side by side
| DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | DeepSeek | Microsoft |
| Noometry Index | 37.8 | 27.9 |
| Released | 2025-01-20 | 2024-04-23 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 13 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 36.8 (#202), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Coding | 51.6% | 15.5% |
| BigCodeBench Instruct | 35.3% | — |
| LMArena Coding | — | 1093 |
| BigCodeBench Complete | 49.9% | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Reasoning DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 24.9 (#156), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Reasoning | 67.6% | 26.8% |
| LiveBench Data Analysis | 55.9% | 34.7% |
| LiveBench | 54.5% | 22.4% |
| Kagi LLM Benchmark | 52.3% | — |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | — | 1072 |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| HellaSwag | — | 76.7% |
| WinoGrande | — | 70.8% |
Math DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 36.0 (#176), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Math | 58.1% | 15.7% |
| OTIS Mock AIME 2024-2025 | 51.4% | — |
| LMArena Math | — | 1111 |
| MATH Level 5 | 89.9% | — |
Knowledge DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 30.7 (#225), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| GPQA Diamond | 55.7% | — |
| LMArena Expert | — | 1045 |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual Not comparable
DeepSeek-R1-Distill-Llama-70B: —, Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | — | 1021 |
| LMArena Chinese | — | 1021 |
| LMArena French | — | 1076 |
| LMArena German | — | 1044 |
| LMArena Japanese | — | 935 |
| LMArena Korean | — | 905 |
| LMArena Russian | — | 1022 |
| LMArena Spanish | — | 1085 |
Instruction Following DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 68.2 (#190), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Instruction Following | 69.9% | 39.1% |
| LMArena Instruction Following | — | 1053 |
Long Context Not comparable
DeepSeek-R1-Distill-Llama-70B: —, Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | — | 1044 |
Writing & Preference DeepSeek-R1-Distill-Llama-70B leads
DeepSeek-R1-Distill-Llama-70B: 49.0 (#194), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | DeepSeek-R1-Distill-Llama-70B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Language | 23.8% | 9.2% |
| LMArena Text | — | 1073 |
| LMArena Creative Writing | — | 1037 |
| LMArena Multi-Turn | — | 1018 |
Frequently asked questions
Is DeepSeek-R1-Distill-Llama-70B better than Phi 3 Mini 4k Instruct?
DeepSeek-R1-Distill-Llama-70B is the stronger model overall, scoring 37.8 to 27.9 on the Noometry Index.
Is DeepSeek-R1-Distill-Llama-70B or Phi 3 Mini 4k Instruct better for coding?
DeepSeek-R1-Distill-Llama-70B scores higher on coding benchmarks: 36.8 versus 26.6 in the Noometry coding category.
How many benchmarks do DeepSeek-R1-Distill-Llama-70B and Phi 3 Mini 4k Instruct share?
7 benchmarks have published results for both models. DeepSeek-R1-Distill-Llama-70B has 13 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.