# Mistral Medium 3

Mistral Medium 3 costs $0.40 per million input tokens and $2.00 per million output tokens, as of October 22, 2026. None of the 6 coding benchmarks behind the index has measured it, so it has no coding index, but 57 benchmarks outside the index have a result below.

Page: https://llmbenchmark.ai/models/mistral-medium-3
Data from October 22, 2026. Every figure below names the board it was read from and the day.

- Maker: Mistral AI
- Released: May 7, 2025
- Price per million input tokens: $0.40
- Price per million output tokens: $2.00
- Context window: 131K tokens

Benchmark results (57 of 195):

- AA Intelligence Index: 9.0 (source: Artificial Analysis (via OpenRouter), read October 22, 2026)
- Design Arena: websites: 1079 Elo (source: Design Arena, read October 22, 2026)
- Design Arena: UI components: 1022 Elo (source: Design Arena, read October 22, 2026)
- Design Arena: game dev: 1017 Elo (source: Design Arena, read October 22, 2026)
- Design Arena: data viz: 1034 Elo (source: Design Arena, read October 22, 2026)
- Design Arena: 3D: 1083 Elo (source: Design Arena, read October 22, 2026)
- Design Arena: all code categories: 1068 Elo (source: Design Arena, read October 22, 2026)
- AA Math Index: 30.3 (source: Artificial Analysis, read October 22, 2026)
- GPQA Diamond (Artificial Analysis run): 57.8% (source: Artificial Analysis, read October 22, 2026)
- Humanity's Last Exam (Artificial Analysis run): 4.1% (source: Artificial Analysis, read October 22, 2026)
- MMLU Pro (Artificial Analysis run): 76.0% (source: Artificial Analysis, read October 22, 2026)
- LiveCodeBench (Artificial Analysis run): 40.0% (source: Artificial Analysis, read October 22, 2026)
- Terminal-Bench Hard (Artificial Analysis run): 3.8% (source: Artificial Analysis, read October 22, 2026)
- IFBench: 39.3% (source: Artificial Analysis, read October 22, 2026)
- Long context reasoning (AA-LCR): 31.3% (source: Artificial Analysis, read October 22, 2026)
- τ²-Bench Telecom: 24.3% (source: Artificial Analysis, read October 22, 2026)
- AIME 2025 (Artificial Analysis run): 30.3% (source: Artificial Analysis, read October 22, 2026)
- DTBench: 62.3% (source: Epoch AI benchmarking hub, read October 22, 2026)
- Epoch Capabilities Index: 134.0 (source: Epoch AI benchmarking hub, read October 22, 2026)
- LMArena Text: 1370 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: coding: 1385 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: hard prompts: 1364 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: math: 1353 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: instruction following: 1338 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: creative writing: 1343 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: 1157 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: English: 1385 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: non-English: 1353 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Chinese: 1374 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: German: 1377 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Japanese: 1316 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Korean: 1304 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: French: 1368 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Spanish: 1364 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Russian: 1359 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: Polish: 1371 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: multi-turn: 1384 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: longer queries: 1360 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: expert: 1342 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: hard prompts in English: 1374 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: software and IT: 1385 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: mathematical work: 1354 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: science: 1381 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: medicine and healthcare: 1381 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: legal and government: 1384 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: business and finance: 1352 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: writing and language: 1352 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Text: entertainment and media: 1335 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: English: 1172 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: Chinese: 1161 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: reading text: 1154 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: diagrams: 1155 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: homework: 1165 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: humor: 1145 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: captioning: 1094 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: recognizing things: 1141 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)
- LMArena Vision: creative writing: 1146 Elo (source: LMArena (Hugging Face dataset), read October 22, 2026)