Model comparison
GPT-4.1 mini vs Phi 3 Mini 4k Instruct
GPT-4.1 mini is the stronger model overall, scoring 33.6 to 27.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
MicrosoftSummary
- They share 18 benchmarks with published results for both. GPT-4.1 mini scores higher in 6 categories and Phi 3 Mini 4k Instruct in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where GPT-4.1 mini leads 73.7 to 47.7.
- The biggest single-benchmark swing is Chess Puzzles: 7% for GPT-4.1 mini 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-4.1 mini | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | OpenAI | Microsoft |
| Noometry Index | 33.6 | 27.9 |
| Released | 2025-04-14 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.40 | — |
| Output $ / M tokens | $1.60 | — |
| Results tracked | 47 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.1 mini leads
GPT-4.1 mini: 30.6 (#293), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1367 | 1093 |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| SciCode | 40.4% | — |
| WeirdML | 37.6% | — |
| BigCodeBench Instruct | 48.9% | — |
| LiveBench Coding | — | 15.5% |
| CadEval | 16% | — |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Agentic & Tool Use Not comparable
GPT-4.1 mini: 33.3 (#55), Phi 3 Mini 4k Instruct: —
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 50.5% | — |
Reasoning Phi 3 Mini 4k Instruct leads
GPT-4.1 mini: 10.8 (#340), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 7% | 0% |
| LMArena Hard Prompts | 1349 | 1072 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 48.6% | — |
| ARC-AGI-1 | 3.5% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | — | 26.8% |
| Mystery Game Puzzles | 7% | — |
| DTBench | 68.8% | — |
| LiveBench Data Analysis | — | 34.7% |
| LMCA | 21.1% | — |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| Epoch Capabilities Index | 135.01 | — |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Phi 3 Mini 4k Instruct leads
GPT-4.1 mini: 24.1 (#270), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1343 | 1111 |
| FrontierMath (Tiers 1-3) | 6.7% | — |
| OTIS Mock AIME 2024-2025 | 44.7% | — |
| Omni-MATH | 49.1% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 87.3% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
Knowledge GPT-4.1 mini leads
GPT-4.1 mini: 34.7 (#194), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1338 | 1045 |
| GPQA Diamond | 65.8% | — |
| SimpleQA Verified | 12.7% | — |
| MMLU-Pro | 78.3% | — |
| GPQA (HELM) | 61.4% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multimodal Not comparable
GPT-4.1 mini: 35.8 (#82), Phi 3 Mini 4k Instruct: —
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Vision | 1181 | — |
Multilingual GPT-4.1 mini leads
GPT-4.1 mini: 45.7 (#166), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1318 | 1021 |
| LMArena Chinese | 1329 | 1021 |
| LMArena French | 1358 | 1076 |
| LMArena German | 1351 | 1044 |
| LMArena Japanese | 1290 | 935 |
| LMArena Korean | 1298 | 905 |
| LMArena Russian | 1324 | 1022 |
| LMArena Spanish | 1319 | 1085 |
Instruction Following GPT-4.1 mini leads
GPT-4.1 mini: 73.7 (#118), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1333 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
| IFEval | 90.4% | — |
Long Context Too close to call
GPT-4.1 mini: 31.8 (#275), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1344 | 1044 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference GPT-4.1 mini leads
GPT-4.1 mini: 48.6 (#199), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | GPT-4.1 mini | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1340 | 1073 |
| LMArena Creative Writing | 1300 | 1037 |
| LMArena Multi-Turn | 1354 | 1018 |
| EQ-Bench Creative Writing | 1147 | — |
| WildBench | 83.8% | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is GPT-4.1 mini better than Phi 3 Mini 4k Instruct?
GPT-4.1 mini is the stronger model overall, scoring 33.6 to 27.9 on the Noometry Index.
Is GPT-4.1 mini or Phi 3 Mini 4k Instruct better for coding?
GPT-4.1 mini scores higher on coding benchmarks: 30.6 versus 26.6 in the Noometry coding category.
How many benchmarks do GPT-4.1 mini and Phi 3 Mini 4k Instruct share?
18 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.