MiniMax M3 vs Mistral Large 4: cost per task
MiniMax M3 is cheaper on 14 of 14 tasks and about 4.1x cheaper on average (geometric mean). On LMArena Text they are practically tied (1441 vs 1432, within the margin of error).
| Task | MiniMax M3 | Mistral Large 4 | Difference |
|---|---|---|---|
| Answer a chat question | 0.12¢ | 0.29¢ | 2.5x |
| Reply to a customer email | 0.07¢ | 0.76¢ | 11.5x |
| Write a 1,500-word blog post | 0.27¢ | 1.6¢ | 6.0x |
| Write 10 product descriptions | 0.21¢ | 0.66¢ | 3.2x |
| Translate 1,000 words | 0.73¢ | 2.6¢ | 3.6x |
| Proofread one page (500 words) | 0.15¢ | 0.92¢ | 5.9x |
| Summarize a 20-page PDF | 0.47¢ | 1.4¢ | 2.9x |
| Extract data from an invoice | 0.12¢ | 0.58¢ | 4.9x |
| Analyze a 200-row spreadsheet | 2.1¢ | 2.6¢ | 1.2x |
| Answer from 5 retrieved documents (RAG) | 0.26¢ | 0.92¢ | 3.5x |
| Classify 100 support tickets | 0.27¢ | 1.6¢ | 6.0x |
| Review 500 lines of code | 0.45¢ | 2.2¢ | 4.9x |
| Write a SQL query | 0.09¢ | 0.67¢ | 7.2x |
| Run a 50-step coding agent | 18¢ | 52¢ | 2.9x |
Specs and list prices
| MiniMax M3 | Mistral Large 4 | |
|---|---|---|
| Input per 1M | $0.30 | $0.68 |
| Output per 1M | $1.20 | $2.09 |
| Cached input per 1M | $0.06 | $0.07 |
| Context | 1M | 1M |
| Providers | 14 | 3 |
| Reasoning | none | default |
| LMArena Text | 1441 | 1432 |
| Released | May 31, 2026 | Oct 6, 2026 |
FAQ
Is MiniMax M3 or Mistral Large 4 cheaper?
MiniMax M3 is cheaper on 14 of 14 tasks and about 4.1x cheaper on average (geometric mean). On LMArena Text they are practically tied (1441 vs 1432, within the margin of error).
How much does MiniMax M3 cost compared to Mistral Large 4 per token?
MiniMax M3: $0.30 input and $1.20 output per 1M tokens. Mistral Large 4: $0.68 input and $2.09 output. Token prices alone mislead - models use different numbers of tokens for the same task, which is what the task table shows.