MiniMax M3 vs Qwen3.8 Max Prime: cost per task
MiniMax M3 is cheaper on 14 of 14 tasks and about 21.8x cheaper on average (geometric mean).
| Task | MiniMax M3 | Qwen3.8 Max Prime | Difference |
|---|---|---|---|
| Answer a chat question | 0.12¢ | 2.0¢ | 16.6x |
| Reply to a customer email | 0.07¢ | 0.95¢ | 14.4x |
| Write a 1,500-word blog post | 0.27¢ | 19¢ | 70.6x |
| Write 10 product descriptions | 0.21¢ | 19¢ | 94.1x |
| Translate 1,000 words | 0.73¢ | 2.1¢ | 2.9x |
| Proofread one page (500 words) | 0.15¢ | 2.5¢ | 16.3x |
| Summarize a 20-page PDF | 0.47¢ | 8.4¢ | 17.8x |
| Extract data from an invoice | 0.12¢ | 2.8¢ | 23.8x |
| Analyze a 200-row spreadsheet | 2.1¢ | 22¢ | 10.5x |
| Answer from 5 retrieved documents (RAG) | 0.26¢ | 5.0¢ | 18.8x |
| Classify 100 support tickets | 0.27¢ | 11¢ | 39.0x |
| Review 500 lines of code | 0.45¢ | 21¢ | 46.6x |
| Write a SQL query | 0.09¢ | 3.2¢ | 34.9x |
| Run a 50-step coding agent | 18¢ | $2.57 | 14.1x |
Specs and list prices
| MiniMax M3 | Qwen3.8 Max Prime | |
|---|---|---|
| Input per 1M | $0.30 | $4.00 |
| Output per 1M | $1.20 | $12.00 |
| Cached input per 1M | $0.06 | $0.50 |
| Context | 1M | 1M |
| Providers | 14 | 1 |
| Reasoning | none | always |
| LMArena Text | 1441 | - |
| Released | May 31, 2026 | Sep 23, 2026 |
FAQ
Is MiniMax M3 or Qwen3.8 Max Prime cheaper?
MiniMax M3 is cheaper on 14 of 14 tasks and about 21.8x cheaper on average (geometric mean).
How much does MiniMax M3 cost compared to Qwen3.8 Max Prime per token?
MiniMax M3: $0.30 input and $1.20 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.