
Phi-3 Mini 128K Instruct
Phi-3 Mini 128K Instruct is an open-weight model from Microsoft. It ranks #396 of 351 on the LMArena text leaderboard with a rating of 1050.
- Developer
- Microsoft
- Released
- Apr 26, 2024
- Context window
- —
- Price in / out, per 1M tokens
- —
- Weights
- Open · MIT
- Inputs
- —
- Reasoning
- —
- Tool use
- —
The rating LMArena published at each update since Apr 26, 2024. Ratings move as votes come in and new models join.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1050 | 396 / 351 | 20.7K |
| CodingLMArena | 1039 | 388 / 346 | 3.9K |
| MathLMArena | 1089 | 363 / 339 | 2.8K |
| Creative writingLMArena | 1024 | 396 / 349 | 2.9K |
| Instruction followingLMArena | 1023 | 394 / 351 | 7.4K |
| Hard promptsLMArena | 1028 | 389 / 351 | 6.2K |
| Multi-turnLMArena | 989 | 398 / 349 | 2.5K |
| Long queriesLMArena | 995 | 392 / 330 | 2.1K |
| ExpertLMArena | 984 | 365 / 305 | 1.1K |
| Legal and governmentIndustries | 1006 | 381 / 325 | 1K |
| Medicine and healthcareIndustries | 996 | 377 / 325 | 1.1K |
| Business and financeIndustries | 990 | 392 / 344 | 2.1K |
| ChineseLanguages | 1016 | 371 / 328 | 2K |
| SpanishLanguages | 1059 | 300 / 249 | 366 |
| GermanLanguages | 1006 | 308 / 261 | 727 |
| FrenchLanguages | 1039 | 295 / 252 | 372 |
| JapaneseLanguages | 899 | 282 / 236 | 385 |
LMArena, CC BY 4.0, as of Oct 22, 2026. Where a model is rated at several effort levels, its best is shown.
Compare Phi-3 Mini 128K Instruct
Microsoft
Microsoft Research builds Phi, small language models that punch above their size. Founded 1975, based in Redmond. azure.microsoft.com
Questions
Who makes Phi-3 Mini 128K Instruct?
Phi-3 Mini 128K Instruct is made by Microsoft, founded 1975 and based in Redmond.
How does Phi-3 Mini 128K Instruct rank?
#396 of 351 on LMArena’s text leaderboard, rated 1050 (95% interval 1043–1058) from 20,685 votes, as of Oct 22, 2026.
Is Phi-3 Mini 128K Instruct open source?
Phi-3 Mini 128K Instruct has open weights under the MIT license.
Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
GPT-5.5
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 Preview