
Phi-3 Medium 4K Instruct
Phi-3 Medium 4K Instruct is an open-weight model from Microsoft. It ranks #348 of 351 on the LMArena text leaderboard with a rating of 1138.
- Developer
- Microsoft
- Released
- Jun 2, 2024
- Context window
- —
- Price in / out, per 1M tokens
- —
- Weights
- Open · MIT
- Inputs
- —
- Reasoning
- —
- Tool use
- —
The rating LMArena published at each update since Jun 2, 2024. Ratings move as votes come in and new models join.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1138 | 348 / 351 | 25.1K |
| CodingLMArena | 1130 | 349 / 346 | 4K |
| MathLMArena | 1173 | 323 / 339 | 3.2K |
| Creative writingLMArena | 1106 | 353 / 349 | 3.7K |
| Instruction followingLMArena | 1113 | 351 / 351 | 9.4K |
| Hard promptsLMArena | 1126 | 346 / 351 | 6.8K |
| Multi-turnLMArena | 1088 | 360 / 349 | 3.8K |
| Long queriesLMArena | 1120 | 350 / 330 | 2.2K |
| ExpertLMArena | 1108 | 331 / 305 | 1.1K |
| Legal and governmentIndustries | 1131 | 342 / 325 | 1.5K |
| Medicine and healthcareIndustries | 1119 | 335 / 325 | 1.3K |
| Business and financeIndustries | 1099 | 347 / 344 | 2.8K |
| ChineseLanguages | 1108 | 344 / 328 | 2.8K |
| SpanishLanguages | 1095 | 288 / 249 | 419 |
| GermanLanguages | 1101 | 285 / 261 | 851 |
| FrenchLanguages | 1111 | 283 / 252 | 350 |
| JapaneseLanguages | 1042 | 259 / 236 | 683 |
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 Medium 4K 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 Medium 4K Instruct?
Phi-3 Medium 4K Instruct is made by Microsoft, founded 1975 and based in Redmond.
How does Phi-3 Medium 4K Instruct rank?
#348 of 351 on LMArena’s text leaderboard, rated 1138 (95% interval 1132–1143) from 25,055 votes, as of Oct 22, 2026.
Is Phi-3 Medium 4K Instruct open source?
Phi-3 Medium 4K 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