GPT-5.4 Mini vs Mistral Large 4: cost per task
GPT-5.4 Mini is cheaper on 12 of 14 tasks and about 1.9x cheaper on average (geometric mean). GPT-5.4 Mini has the higher LMArena Text score (1447 vs 1432).
| Task | GPT-5.4 Mini | Mistral Large 4 | Difference |
|---|---|---|---|
| Answer a chat question | 0.53¢ | 0.29¢ | 1.8x |
| Reply to a customer email | 0.06¢ | 0.76¢ | 12.3x |
| Write a 1,500-word blog post | 1.1¢ | 1.6¢ | 1.5x |
| Write 10 product descriptions | 0.48¢ | 0.66¢ | 1.4x |
| Translate 1,000 words | 0.73¢ | 2.6¢ | 3.6x |
| Proofread one page (500 words) | 0.39¢ | 0.92¢ | 2.3x |
| Summarize a 20-page PDF | 1.2¢ | 1.4¢ | 1.1x |
| Extract data from an invoice | 0.28¢ | 0.58¢ | 2.1x |
| Analyze a 200-row spreadsheet | 1.6¢ | 2.6¢ | 1.6x |
| Answer from 5 retrieved documents (RAG) | 0.43¢ | 0.92¢ | 2.2x |
| Classify 100 support tickets | 0.56¢ | 1.6¢ | 2.9x |
| Review 500 lines of code | 2.7¢ | 2.2¢ | 1.2x |
| Write a SQL query | 0.36¢ | 0.67¢ | 1.8x |
| Run a 50-step coding agent | 32¢ | 52¢ | 1.6x |
Specs and list prices
| GPT-5.4 Mini | Mistral Large 4 | |
|---|---|---|
| Input per 1M | $0.75 | $0.68 |
| Output per 1M | $4.50 | $2.09 |
| Cached input per 1M | $0.075 | $0.07 |
| Context | 400K | 1M |
| Providers | 5 | 3 |
| Reasoning | default | default |
| LMArena Text | 1447 | 1432 |
| Released | Mar 17, 2026 | Oct 6, 2026 |
FAQ
Is GPT-5.4 Mini or Mistral Large 4 cheaper?
GPT-5.4 Mini is cheaper on 12 of 14 tasks and about 1.9x cheaper on average (geometric mean). GPT-5.4 Mini has the higher LMArena Text score (1447 vs 1432).
How much does GPT-5.4 Mini cost compared to Mistral Large 4 per token?
GPT-5.4 Mini: $0.75 input and $4.50 output per 1M tokens. Mistral Large 4: $0.68 input and $2.09 output. Token prices alone mislead - models use different numbers of tokens for the same task, which is what the task table shows.