Model comparison
GPT-5 Nano vs Phi 3 Mini 4k Instruct
GPT-5 Nano is the stronger model overall, scoring 33.5 to 27.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
MicrosoftSummary
- They share 17 benchmarks with published results for both. GPT-5 Nano scores higher in 7 categories and Phi 3 Mini 4k Instruct in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where GPT-5 Nano leads 75.0 to 47.7.
- The biggest single-benchmark swing is Chess Puzzles: 27% for GPT-5 Nano and 0% for Phi 3 Mini 4k Instruct.
- Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Nano | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 33.5 | 27.9 |
| Released | 2025-08-07 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.40 | — |
| Results tracked | 49 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Nano leads
GPT-5 Nano: 33.6 (#254), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1351 | 1093 |
| SWE-bench Verified (bash only) | 34.8% | — |
| WeirdML | 38.1% | — |
| LiveBench Coding | — | 15.5% |
| ALE-Bench | 718.67 | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
GPT-5 Nano: 25.8 (#106), Phi 3 Mini 4k Instruct: —
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| Terminal-Bench | 21.8% | — |
| Berkeley Function Calling Leaderboard | 51.5% | — |
Reasoning GPT-5 Nano leads
GPT-5 Nano: 16.3 (#306), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 27% | 0% |
| LMArena Hard Prompts | 1328 | 1072 |
| ARC-AGI-2 | 2.6% | — |
| Kagi LLM Benchmark | 62.2% | — |
| ARC-AGI-1 | 20.7% | — |
| LiveBench Reasoning | — | 26.8% |
| Mystery Game Puzzles | 9% | — |
| DTBench | 62.7% | — |
| LiveBench Data Analysis | — | 34.7% |
| LMCA | 7.9% | — |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 139.38 | — |
| ForecastBench | 59.1 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math GPT-5 Nano leads
GPT-5 Nano: 29.4 (#241), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1317 | 1111 |
| FrontierMath (Tiers 1-3) | 20% | — |
| FrontierMath Tier 4 | 2.4% | — |
| OTIS Mock AIME 2024-2025 | 81.1% | — |
| ProofBench | 12% | — |
| Omni-MATH | 54.6% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 95.2% | — |
| FrontierMath (Feb 2025 set) | 8.3% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge GPT-5 Nano leads
GPT-5 Nano: 35.9 (#178), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1321 | 1045 |
| GPQA Diamond | 69.4% | — |
| SimpleQA Verified | 11.7% | — |
| MMLU-Pro | 77.8% | — |
| Vectara Hallucination Rate | 10.5% | — |
| GPQA (HELM) | 67.9% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multimodal Not comparable
GPT-5 Nano: 31.3 (#108), Phi 3 Mini 4k Instruct: —
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Vision | 1159 | — |
| VPCT | 37.2% | — |
Multilingual GPT-5 Nano leads
GPT-5 Nano: 45.3 (#172), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1313 | 1021 |
| LMArena Chinese | 1356 | 1021 |
| LMArena German | 1327 | 1044 |
| LMArena Japanese | 1226 | 935 |
| LMArena Korean | 1269 | 905 |
| LMArena Russian | 1296 | 1022 |
| LMArena Spanish | 1360 | 1085 |
| LMArena French | — | 1076 |
Instruction Following GPT-5 Nano leads
GPT-5 Nano: 75.0 (#79), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1306 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
| IFEval | 93.2% | — |
Long Context Too close to call
GPT-5 Nano: 31.3 (#281), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1312 | 1044 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference GPT-5 Nano leads
GPT-5 Nano: 39.1 (#249), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | GPT-5 Nano | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1320 | 1073 |
| LMArena Creative Writing | 1249 | 1037 |
| LMArena Multi-Turn | 1311 | 1018 |
| EQ-Bench Creative Writing | 705 | — |
| WildBench | 80.6% | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is GPT-5 Nano better than Phi 3 Mini 4k Instruct?
GPT-5 Nano is the stronger model overall, scoring 33.5 to 27.9 on the Noometry Index.
Is GPT-5 Nano or Phi 3 Mini 4k Instruct better for coding?
GPT-5 Nano scores higher on coding benchmarks: 33.6 versus 26.6 in the Noometry coding category.
How many benchmarks do GPT-5 Nano and Phi 3 Mini 4k Instruct share?
17 benchmarks have published results for both models. GPT-5 Nano has 49 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.