MicrosoftPhi 3 Mini
Phi 3 Mini is a Microsoft open-weights model, released 23 Apr 2024. Its best configuration ranks #743 of 785 on the BenchLeader Index at 36.8 ±2.5, in the lower half. The ± is the point: 162 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in composite (36) and lowest in human preference (26). Last measured 22 Oct 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- 4k
- Full answer
- –
- Cost per run
- –
- Released
- 23 Apr 2024
- Overall index37
- Reasoning30
- Coding29
- Agents & tools34
- Maths35
- Instruction following29
- Human preference26
- Composite36
Versions
Microsoft has shipped 3 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Phi-4-Mini | 27 Feb 2025 | 35.8 | #761 |
| Phi-3.5-mini | 16 Aug 2024 | – | – |
| Phi 3 Minithis page | 23 Apr 2024 | 36.8 | #743 |
How its index has moved
9 Sept 2026 to 23 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1176#327 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 31.9%#513 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 5.0%#459 | Artificial Analysis | |
| HellaSwagnot in index | 76.7%#36 | Epoch AI Benchmarking Hub | |
| BIG-Bench Hardnot in index | 71.7%#10 | Epoch AI Benchmarking Hub | |
| WinoGrandenot in index | 70.8%#44 | Epoch AI Benchmarking Hub | |
| ARC (AI2 Challenge)not in index | 84.9%#14 | Epoch AI Benchmarking Hub | |
| ANLInot in index | 52.8%#6 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Coding | 1197#327 | Arena (LMArena) |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Terminal-Bench Hard | 0.0%#362 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 0.0%#393 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Maths | 1194#302 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| MMLUnot in index | 68.8%#59 | Epoch AI Benchmarking Hub | |
| TriviaQAnot in index | 64.0%#32 | Epoch AI Benchmarking Hub | |
| OpenBookQAnot in index | 88.0%#1 | Epoch AI Benchmarking Hub |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| IFBench | 23.9%#398 | Artificial Analysis | |
| Arena Instruction Followingnot in index | 1125#336 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1143#341 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1096#345 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1099#340 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1109#343 | Arena (LMArena) |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Intelligence Index v4.3.2 | 5.8#556 | Artificial Analysis |
More from Microsoft
See also
Questions people ask
- What are Phi 3 Mini's benchmark scores?
- As of 25 Oct 2026, Phi 3 Mini scores 36.8 ±2.5 on the BenchLeader Index, strongest in composite (36) and weakest in human preference (26). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does Phi 3 Mini rank?
- Phi 3 Mini is #743 of 785 ranked configurations on 25 Oct 2026. 162 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- When was Phi 3 Mini released?
- Microsoft released Phi 3 Mini on 23 Apr 2024.
Data as of 25 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Phi 3 Mini: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/phi-3-mini.