GPT-5.4 Nano vs Qwen3.8 Max Prime: cost per task
GPT-5.4 Nano is cheaper on 14 of 14 tasks and about 36.0x cheaper on average (geometric mean).
| Task | GPT-5.4 Nano | Qwen3.8 Max Prime | Difference |
|---|---|---|---|
| Answer a chat question | 0.15¢ | 2.0¢ | 13.2x |
| Reply to a customer email | 0.02¢ | 0.95¢ | 56.3x |
| Write a 1,500-word blog post | 0.30¢ | 19¢ | 65.0x |
| Write 10 product descriptions | 0.13¢ | 19¢ | 145.1x |
| Translate 1,000 words | 0.20¢ | 2.1¢ | 10.5x |
| Proofread one page (500 words) | 0.11¢ | 2.5¢ | 23.1x |
| Summarize a 20-page PDF | 0.33¢ | 8.4¢ | 25.8x |
| Extract data from an invoice | 0.08¢ | 2.8¢ | 36.2x |
| Analyze a 200-row spreadsheet | 0.44¢ | 22¢ | 48.9x |
| Answer from 5 retrieved documents (RAG) | 0.12¢ | 5.0¢ | 42.8x |
| Classify 100 support tickets | 0.15¢ | 11¢ | 69.1x |
| Review 500 lines of code | 0.75¢ | 21¢ | 28.3x |
| Write a SQL query | 0.10¢ | 3.2¢ | 32.2x |
| Run a 50-step coding agent | 8.7¢ | $2.57 | 29.6x |
Specs and list prices
| GPT-5.4 Nano | Qwen3.8 Max Prime | |
|---|---|---|
| Input per 1M | $0.20 | $4.00 |
| Output per 1M | $1.25 | $12.00 |
| Cached input per 1M | $0.02 | $0.50 |
| Context | 400K | 1M |
| Providers | 4 | 1 |
| Reasoning | default | always |
| LMArena Text | 1401 | - |
| Released | Mar 17, 2026 | Sep 23, 2026 |
FAQ
Is GPT-5.4 Nano or Qwen3.8 Max Prime cheaper?
GPT-5.4 Nano is cheaper on 14 of 14 tasks and about 36.0x cheaper on average (geometric mean).
How much does GPT-5.4 Nano cost compared to Qwen3.8 Max Prime per token?
GPT-5.4 Nano: $0.20 input and $1.25 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.