Mistral Small 4 vs Qwen3.8 Max Prime: cost per task
Mistral Small 4 is cheaper on 14 of 14 tasks and about 19.0x cheaper on average (geometric mean).
| Task | Mistral Small 4 | Qwen3.8 Max Prime | Difference |
|---|---|---|---|
| Answer a chat question | 0.08¢ | 2.0¢ | 23.5x |
| Reply to a customer email | 0.22¢ | 0.95¢ | 4.4x |
| Write a 1,500-word blog post | 0.47¢ | 19¢ | 41.2x |
| Write 10 product descriptions | 0.19¢ | 19¢ | 102.3x |
| Translate 1,000 words | 0.75¢ | 2.1¢ | 2.8x |
| Proofread one page (500 words) | 0.26¢ | 2.5¢ | 9.7x |
| Summarize a 20-page PDF | 0.33¢ | 8.4¢ | 25.4x |
| Extract data from an invoice | 0.16¢ | 2.8¢ | 17.1x |
| Analyze a 200-row spreadsheet | 0.71¢ | 22¢ | 30.5x |
| Answer from 5 retrieved documents (RAG) | 0.25¢ | 5.0¢ | 20.0x |
| Classify 100 support tickets | 0.46¢ | 11¢ | 23.1x |
| Review 500 lines of code | 0.62¢ | 21¢ | 34.3x |
| Write a SQL query | 0.19¢ | 3.2¢ | 16.9x |
| Run a 50-step coding agent | 13¢ | $2.57 | 19.1x |
Specs and list prices
| Mistral Small 4 | Qwen3.8 Max Prime | |
|---|---|---|
| Input per 1M | $0.15 | $4.00 |
| Output per 1M | $0.60 | $12.00 |
| Cached input per 1M | $0.015 | $0.50 |
| Context | 262K | 1M |
| Providers | 4 | 1 |
| Reasoning | default | always |
| LMArena Text | - | - |
| Released | Mar 16, 2026 | Sep 23, 2026 |
FAQ
Is Mistral Small 4 or Qwen3.8 Max Prime cheaper?
Mistral Small 4 is cheaper on 14 of 14 tasks and about 19.0x cheaper on average (geometric mean).
How much does Mistral Small 4 cost compared to Qwen3.8 Max Prime per token?
Mistral Small 4: $0.15 input and $0.60 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.