MicrosoftPhi 3
As of 6 Nov 2026, Phi 3 by Microsoft scores 37.9 on the BenchLeader Index (587 ranked configurations), with its best results in maths (35) and weakest in reasoning (26). Released 23 Apr 2024.
Key facts (as of 6 Nov 2026)
- Phi 3 was released on 23 Apr 2024 (Microsoft).
In detail. Phi 3 is a Microsoft model, released 23 Apr 2024. Its best configuration (medium reasoning effort) ranks #547 of 587 on the BenchLeader Index at 37.9 ±3.6, in the lower half. The ± is the point: 124 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 maths (35) and lowest in reasoning (26). Last measured 28 Oct 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- –
- Full answer
- –
- Cost per task
- –
- Tokens per task
- –
- Released
- 23 Apr 2024
- Overall index38
- Reasoning26
- Coding33
- Maths35
- Human preference33
Versions
Microsoft has shipped 4 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.
How its index has moved
27 Sep 2026 to 27 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 | medium | Source | Trend |
|---|---|---|---|
| GPQA Diamond | 27.6%#285 | Epoch AI Benchmarking Hub | |
| Arena Hard Prompts | 1213#315 | Arena (LMArena) | |
| HellaSwagnot in index | 82.4%#19 | Epoch AI Benchmarking Hub | |
| BIG-Bench Hardnot in index | 81.4%#5 | Epoch AI Benchmarking Hub | |
| WinoGrandenot in index | 81.5%#14 | Epoch AI Benchmarking Hub | |
| ARC (AI2 Challenge)not in index | 91.6%#5 | Epoch AI Benchmarking Hub | |
| ANLInot in index | 55.8%#4 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Coding | 1231#318 | Arena (LMArena) |
Maths
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| MATH Level 5 | 17.6%#81 | Epoch AI Benchmarking Hub | |
| Arena Maths | 1216#288 | Arena (LMArena) |
Knowledge
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| MMLUnot in index | 78.0%#28 | Epoch AI Benchmarking Hub | |
| TriviaQAnot in index | 73.9%#23 | Epoch AI Benchmarking Hub | |
| OpenBookQAnot in index | 87.4%#3 | Epoch AI Benchmarking Hub |
Instruction following
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Instruction Followingnot in index | 1178#315 | Arena (LMArena) |
Human preference
| Benchmark | medium | Source | Trend |
|---|---|---|---|
| Arena Text | 1198#314 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1150#319 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1140#325 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1192#314 | Arena (LMArena) |
More from Microsoft
See also
Embed or cite
Show where Phi 3 ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.
[](https://www.benchleader.com/models/phi-3)Cite this
The ranking changes daily, so a reference carries the data date. APA style:
BenchLeader. (2026, November 6). Phi 3 on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 6, 2026, from https://www.benchleader.com/models/phi-3
BibTeX:
@misc{benchleader_phi_3_20261106,
author = {{BenchLeader}},
title = {{Phi 3 on the BenchLeader Index (method v1.4)}},
year = {2026},
month = nov,
howpublished = {\url{https://www.benchleader.com/models/phi-3}},
note = {Data as of 2026-11-06}
}Every badge style, the embeddable top-10 widget and how to cite the Index
Questions people ask
- What are Phi 3's benchmark scores?
- As of 6 Nov 2026, Phi 3 (medium effort) scores 37.9 ±3.6 on the BenchLeader Index, strongest in maths (35) and weakest in reasoning (26). That rests on independent benchmark results only, and every score on this page links to its publisher.
- How does Phi 3 rank?
- Phi 3 (medium effort) is #547 of 587 ranked configurations on 6 Nov 2026. 124 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 69.3.
- When was Phi 3 released?
- Microsoft released Phi 3 on 23 Apr 2024.
Data as of 6 Nov 2026. Compare with another model.
Cite as: BenchLeader, “Phi 3: benchmarks, pricing, speed and rank”, data as of 6 Nov 2026, https://www.benchleader.com/models/phi-3.