Mistral Large 4 vs Qwen3.8 Max Prime: cost per task
Mistral Large 4 is cheaper on 13 of 14 tasks and about 5.3x cheaper on average (geometric mean).
| Task | Mistral Large 4 | Qwen3.8 Max Prime | Difference |
|---|---|---|---|
| Answer a chat question | 0.29¢ | 2.0¢ | 6.7x |
| Reply to a customer email | 0.76¢ | 0.95¢ | 1.3x |
| Write a 1,500-word blog post | 1.6¢ | 19¢ | 11.8x |
| Write 10 product descriptions | 0.66¢ | 19¢ | 29.2x |
| Translate 1,000 words | 2.6¢ | 2.1¢ | 1.2x |
| Proofread one page (500 words) | 0.92¢ | 2.5¢ | 2.7x |
| Summarize a 20-page PDF | 1.4¢ | 8.4¢ | 6.2x |
| Extract data from an invoice | 0.58¢ | 2.8¢ | 4.8x |
| Analyze a 200-row spreadsheet | 2.6¢ | 22¢ | 8.5x |
| Answer from 5 retrieved documents (RAG) | 0.92¢ | 5.0¢ | 5.4x |
| Classify 100 support tickets | 1.6¢ | 11¢ | 6.5x |
| Review 500 lines of code | 2.2¢ | 21¢ | 9.5x |
| Write a SQL query | 0.67¢ | 3.2¢ | 4.8x |
| Run a 50-step coding agent | 52¢ | $2.57 | 4.9x |
Specs and list prices
| Mistral Large 4 | Qwen3.8 Max Prime | |
|---|---|---|
| Input per 1M | $0.68 | $4.00 |
| Output per 1M | $2.09 | $12.00 |
| Cached input per 1M | $0.07 | $0.50 |
| Context | 1M | 1M |
| Providers | 3 | 1 |
| Reasoning | default | always |
| LMArena Text | 1432 | - |
| Released | Oct 6, 2026 | Sep 23, 2026 |
FAQ
Is Mistral Large 4 or Qwen3.8 Max Prime cheaper?
Mistral Large 4 is cheaper on 13 of 14 tasks and about 5.3x cheaper on average (geometric mean).
How much does Mistral Large 4 cost compared to Qwen3.8 Max Prime per token?
Mistral Large 4: $0.68 input and $2.09 output per 1M tokens. Qwen3.8 Max Prime: $4.00 input and $12.00 output. Token prices alone mislead - models use different numbers of tokens for the same task, which is what the task table shows.