Llama 4 Maverick vs Qwen3.8 Max Prime: cost per task
Llama 4 Maverick is cheaper on 14 of 14 tasks and about 18.4x cheaper on average (geometric mean).
| Task | Llama 4 Maverick | Qwen3.8 Max Prime | Difference |
|---|---|---|---|
| Answer a chat question | 0.11¢ | 2.0¢ | 18.5x |
| Reply to a customer email | 0.05¢ | 0.95¢ | 18.8x |
| Write a 1,500-word blog post | 1.0¢ | 19¢ | 18.4x |
| Write 10 product descriptions | 1.1¢ | 19¢ | 18.4x |
| Translate 1,000 words | 0.11¢ | 2.1¢ | 19.2x |
| Proofread one page (500 words) | 0.13¢ | 2.5¢ | 18.8x |
| Summarize a 20-page PDF | 0.42¢ | 8.4¢ | 20.2x |
| Extract data from an invoice | 0.15¢ | 2.8¢ | 18.6x |
| Analyze a 200-row spreadsheet | 1.2¢ | 22¢ | 18.7x |
| Answer from 5 retrieved documents (RAG) | 0.26¢ | 5.0¢ | 19.3x |
| Classify 100 support tickets | 0.58¢ | 11¢ | 18.5x |
| Review 500 lines of code | 1.1¢ | 21¢ | 18.6x |
| Write a SQL query | 0.17¢ | 3.2¢ | 18.5x |
| Run a 50-step coding agent | 19¢ | $2.57 | 13.8x |
Specs and list prices
| Llama 4 Maverick | Qwen3.8 Max Prime | |
|---|---|---|
| Input per 1M | $0.19 | $4.00 |
| Output per 1M | $0.65 | $12.00 |
| Cached input per 1M | $0.05 | $0.50 |
| Context | 1M | 1M |
| Providers | 4 | 1 |
| Reasoning | none | always |
| LMArena Text | 1327 | - |
| Released | Apr 5, 2025 | Sep 23, 2026 |
FAQ
Is Llama 4 Maverick or Qwen3.8 Max Prime cheaper?
Llama 4 Maverick is cheaper on 14 of 14 tasks and about 18.4x cheaper on average (geometric mean).
How much does Llama 4 Maverick cost compared to Qwen3.8 Max Prime per token?
Llama 4 Maverick: $0.19 input and $0.65 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.