MiniMax M3 API pricing
MiniMax's models. MiniMax M3 costs $0.30 / $1.20 per 1M tokens up to 512K tokens of context, and $0.60 / $2.40 above.
MiniMax M3 prices
What it costs in practice
MiniMax M3 at list prices, no batch discount.
| 1,000 chat replies1,500 tokens in and 300 out each | $0.81 |
|---|---|
| Summarising 100 long reports20,000 tokens in and 800 out each | $0.70 |
| A month of a coding agent400 tasks, 85% of input cached | $81.60 |
The coding-agent example assumes 85% of input is served from cache, typical for agents that resend the same context each step. Model your own workload
Details
MiniMax-M3- Prompts above 512K tokens cost $0.60 / $2.40 per 1M.
Checked 27 Sep 2026 against MiniMax pricing .
Compare MiniMax M3
No ready-made comparisons for this family yet.
Price history
All changesNo price changes recorded since we began tracking in June 2026.
MiniMax pricing rules
- MiniMax describes its M3 rates as a permanent 50% discount.
- The Priority tier costs 1.5× the standard price.
Questions
How much does MiniMax M3 cost?
As of 27 Sep 2026, MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens, and $0.06 per million cached input tokens. Prices checked 27 Sep 2026.
What is MiniMax M3's context window?
1M tokens.
Does MiniMax M3 support prompt caching?
Yes. Repeated input read from the cache costs $0.06 per million tokens, 80% less than normal input.