Model comparison
o1-mini vs Phi 3 Mini 4k Instruct
o1-mini is the stronger model overall, scoring 34.0 to 27.9 on the Noometry Index.
Last verified . 26 shared benchmarks.
MicrosoftSummary
- They share 26 benchmarks with published results for both. o1-mini scores higher in 7 categories and Phi 3 Mini 4k Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where o1-mini leads 48.4 to 27.6.
- The biggest single-benchmark swing is LiveBench Math: 62% for o1-mini and 15.7% for Phi 3 Mini 4k Instruct.
- Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| o1-mini | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 34.0 | 27.9 |
| Released | 2024-09-12 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 39 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1-mini leads
o1-mini: 35.5 (#224), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Coding | 48% | 15.5% |
| LMArena Coding | 1362 | 1093 |
| HumanEval+ | 89% | 59.1% |
| MBPP+ | 78.8% | 54.2% |
| Aider Polyglot | 32.9% | — |
| WeirdML | 36.3% | — |
Agentic & Tool Use Not comparable
o1-mini: 24.6 (#118), Phi 3 Mini 4k Instruct: —
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| Cybench | 10% | — |
Reasoning Phi 3 Mini 4k Instruct leads
o1-mini: 8.8 (#346), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Reasoning | 72.3% | 26.8% |
| LMArena Hard Prompts | 1333 | 1072 |
| LiveBench Data Analysis | 57.9% | 34.7% |
| LiveBench | 57.8% | 22.4% |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 18.1% | — |
| ARC-AGI-1 | 14% | — |
| Chess Puzzles | — | 0% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 135.82 | — |
| HellaSwag | — | 76.7% |
| WinoGrande | — | 70.8% |
Math o1-mini leads
o1-mini: 35.4 (#186), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Math | 62% | 15.7% |
| LMArena Math | 1358 | 1111 |
| OTIS Mock AIME 2024-2025 | 46.9% | — |
| MATH Level 5 | 89.2% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge o1-mini leads
o1-mini: 34.9 (#192), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1316 | 1045 |
| GPQA Diamond | 62.4% | — |
| Confabulations | 18.6% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual o1-mini leads
o1-mini: 43.6 (#182), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1289 | 1021 |
| LMArena Chinese | 1314 | 1021 |
| LMArena French | 1293 | 1076 |
| LMArena German | 1278 | 1044 |
| LMArena Japanese | 1245 | 935 |
| LMArena Korean | 1223 | 905 |
| LMArena Russian | 1283 | 1022 |
| LMArena Spanish | 1303 | 1085 |
Instruction Following o1-mini leads
o1-mini: 66.7 (#206), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LiveBench Instruction Following | 65.4% | 39.1% |
| LMArena Instruction Following | 1304 | 1053 |
Long Context o1-mini leads
o1-mini: 40.1 (#161), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1320 | 1044 |
Writing & Preference o1-mini leads
o1-mini: 48.4 (#202), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | o1-mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1317 | 1073 |
| LMArena Creative Writing | 1244 | 1037 |
| LMArena Multi-Turn | 1314 | 1018 |
| LiveBench Language | 40.9% | 9.2% |
| Short-Story Creative Writing | 64.9% | — |
Frequently asked questions
Is o1-mini better than Phi 3 Mini 4k Instruct?
o1-mini is the stronger model overall, scoring 34.0 to 27.9 on the Noometry Index.
Is o1-mini or Phi 3 Mini 4k Instruct better for coding?
o1-mini scores higher on coding benchmarks: 35.5 versus 26.6 in the Noometry coding category.
How many benchmarks do o1-mini and Phi 3 Mini 4k Instruct share?
26 benchmarks have published results for both models. o1-mini has 39 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.