Model comparison
GPT-4o mini vs Phi 3 Small 8k Instruct
Phi 3 Small 8k Instruct is the stronger model overall, scoring 28.8 to 25.4 on the Noometry Index.
Last verified . 25 shared benchmarks.
MicrosoftSummary
- They share 25 benchmarks with published results for both. GPT-4o mini scores higher in 4 categories and Phi 3 Small 8k Instruct in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Phi 3 Small 8k Instruct leads 25.4 to 9.9.
- The biggest single-benchmark swing is LiveBench Coding: 43.1% for GPT-4o mini and 20.3% for Phi 3 Small 8k Instruct.
- Phi 3 Small 8k Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | Phi 3 Small 8k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 25.4 | 28.8 |
| Released | 2024-07-18 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 60 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi 3 Small 8k Instruct leads
GPT-4o mini: 21.6 (#330), Phi 3 Small 8k Instruct: 27.4 (#313)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Coding | 43.1% | 20.3% |
| LMArena Coding | 1290 | 1102 |
| Aider Polyglot | 3.6% | — |
| WeirdML | 11.8% | — |
| BigCodeBench Instruct | 46.1% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Not comparable
GPT-4o mini: 27.2 (#95), Phi 3 Small 8k Instruct: —
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| BALROG | 17.4% | — |
Reasoning Phi 3 Small 8k Instruct leads
GPT-4o mini: 8.7 (#343), Phi 3 Small 8k Instruct: 14.2 (#319)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Reasoning | 32.8% | 15.9% |
| LMArena Hard Prompts | 1267 | 1100 |
| LiveBench Data Analysis | 50% | 30.3% |
| LiveBench | 41.3% | 24% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 10.7% | — |
| Kagi LLM Benchmark | 28.8% | — |
| Chess Puzzles | 0% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 54.4% | — |
| LMCA | 10.4% | — |
| Adversarial NLI | — | 58.1% |
| BIG-Bench Hard | — | 79.1% |
| Epoch Capabilities Index | 126.56 | — |
| HellaSwag | — | 77% |
| PIQA | 88.7% | — |
| WinoGrande | — | 81.5% |
Math Phi 3 Small 8k Instruct leads
GPT-4o mini: 9.9 (#307), Phi 3 Small 8k Instruct: 25.4 (#247)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Math | 36.3% | 17.6% |
| LMArena Math | 1267 | 1151 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| OTIS Mock AIME 2024-2025 | 6.9% | — |
| Omni-MATH | 28% | — |
| MATH Level 5 | 52.6% | — |
| GSM8K | 91.3% | — |
Knowledge Phi 3 Small 8k Instruct leads
GPT-4o mini: 17.7 (#279), Phi 3 Small 8k Instruct: 28.5 (#237)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Expert | 1235 | 1067 |
| MMLU | 81.8% | 75.7% |
| GPQA Diamond | 37.7% | — |
| SimpleQA Verified | 8.3% | — |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| GPQA (HELM) | 36.8% | — |
| ARC (AI2) Challenge | — | 90.7% |
| BoolQ | 88.7% | — |
| OpenBookQA | — | 88% |
| TriviaQA | — | 58.1% |
Multimodal Not comparable
GPT-4o mini: 25.3 (#117), Phi 3 Small 8k Instruct: —
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Vision | 1066 | — |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
Multilingual GPT-4o mini leads
GPT-4o mini: 42.2 (#194), Phi 3 Small 8k Instruct: 28.6 (#267)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Non-English | 1266 | 1058 |
| LMArena Chinese | 1264 | 1061 |
| LMArena French | 1296 | 1134 |
| LMArena German | 1272 | 1080 |
| LMArena Japanese | 1216 | 966 |
| LMArena Korean | 1195 | 894 |
| LMArena Russian | 1275 | 1111 |
| LMArena Spanish | 1276 | 1111 |
Instruction Following GPT-4o mini leads
GPT-4o mini: 62.1 (#234), Phi 3 Small 8k Instruct: 52.2 (#287)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LiveBench Instruction Following | 56.8% | 47.2% |
| LMArena Instruction Following | 1259 | 1087 |
| IFEval | 78.2% | — |
Long Context GPT-4o mini leads
GPT-4o mini: 39.1 (#181), Phi 3 Small 8k Instruct: 33.0 (#262)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Longer Query | 1289 | 1088 |
Writing & Preference GPT-4o mini leads
GPT-4o mini: 39.7 (#243), Phi 3 Small 8k Instruct: 31.3 (#279)
| Benchmark | GPT-4o mini | Phi 3 Small 8k Instruct |
|---|---|---|
| LMArena Text | 1286 | 1110 |
| LMArena Creative Writing | 1268 | 1083 |
| LMArena Multi-Turn | 1285 | 1068 |
| LiveBench Language | 28.6% | 12.9% |
| Short-Story Creative Writing | 67.2% | — |
| EQ-Bench Creative Writing | 873 | — |
| WildBench | 79.1% | — |
Frequently asked questions
Is GPT-4o mini better than Phi 3 Small 8k Instruct?
Phi 3 Small 8k Instruct is the stronger model overall, scoring 28.8 to 25.4 on the Noometry Index.
Is GPT-4o mini or Phi 3 Small 8k Instruct better for coding?
Phi 3 Small 8k Instruct scores higher on coding benchmarks: 27.4 versus 21.6 in the Noometry coding category.
How many benchmarks do GPT-4o mini and Phi 3 Small 8k Instruct share?
25 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Phi 3 Small 8k Instruct has 32.