Model tracker / MiniMax
MiniMax M3
MiniMax M3 is an everyday open-weight model from MiniMax, released 1 June 2026. At MiniMax it costs $0.30 per million input tokens and $1.20 per million output tokens, with a 1M-token context window. 6 providers serve it; the cheapest is Fireworks AI at $0.30 input and $1.20 output.
- Made by
- MiniMax
- Released
- 2026-06-01
- Status
- Generally available
- Tier
- Everyday
- Weights
- Open
- Context window
- 1,000,000 tokens
- Input
- text, image
- API id
MiniMax-M3
MiniMax M3 price at 6 providers
US dollars per million tokens. Blended is 3 input tokens for every output token, one number to compare on. The cheapest blended price is highlighted.
| Provider | Input | Output | Cached input | Blended | Batch | Checked |
|---|---|---|---|---|---|---|
| Fireworks AI · accounts/fireworks/models/minimax-m3 | $0.30 | $1.20 | $0.06 | $0.525 | 50% off | 2026-10-14 |
| MiniMax · maker | $0.30 | $1.20 | $0.06 | $0.525 | – | 2026-10-14 |
| Nebius Group N.V. · MiniMaxAI/MiniMax-M3 | $0.30 | $1.20 | – | $0.525 | – | 2026-10-14 |
| Novita AI · minimax/minimax-m3 | $0.30 | $1.20 | $0.06 | $0.525 | – | 2026-10-14 |
| Together AI · MiniMaxAI/MiniMax-M3 | $0.30 | $1.20 | $0.06 | $0.525 | – | 2026-10-14 |
| SambaNova | $0.60 | $2.40 | $0.06 | $1.05 | – | 2026-10-14 |
- Fireworks AI: Priority tier: $0.45 / $1.80.
- MiniMax: Up to 512K input tokens, shown as a permanent 50% off $0.60 / $2.40. Above 512K: $0.60 in / $0.12 cached / $2.40 out. Priority tier 1.5x. No batch price.
- Nebius Group N.V.: No cached-input or batch price published.
- Novita AI: Up to 512K input tokens. 512K-1M: $0.60 / $2.40, cache read $0.12.
- SambaNova: Preview model on SambaCloud; text-only deployment.
MiniMax M3 benchmarks
| Benchmark | Score | What it measures | Reported by |
|---|---|---|---|
| LMArena text score | 1440 | Elo-style score from people voting between answers | Independent |
Scores reported by a model's maker aren't independently checked.
Models like MiniMax M3
Other everyday models closest in price, at each maker's own price or the cheapest provider.
| Model | Input | Output | Context | Weights |
|---|---|---|---|---|
| MiniMax M3 | $0.30 | $1.20 | 1M | Open |
| Qwen3.7 Plus Alibaba | $0.40 | $1.60 | 1M | Closed |
| Mistral Large 3 Mistral AI | $0.50 | $1.50 | 256K | Open |
| NVIDIA Nemotron 3 Ultra NVIDIA | $0.50 | $2.20 | 256K | Open |
| Step 5 Preview StepFun | $1.00 | $2.70 | 1M | Closed |
| DeepSeek V4 Pro DeepSeek | $1.32 | $3.96 | 1M | Open |
Compare them all in the model tracker →
MiniMax M3 FAQ
How much does MiniMax M3 cost?
MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens at MiniMax, and $0.06 per million cached input tokens. Prices are list prices in US dollars, last checked 14 October 2026.
What is MiniMax M3's context window?
MiniMax M3 takes up to 1,000,000 tokens in one request.
Is MiniMax M3 open weight?
Yes. MiniMax publishes the weights, so you can run MiniMax M3 yourself or at any host that serves it.
Where can I use MiniMax M3?
MiniMax M3 is available from Fireworks AI, MiniMax, Nebius Group N.V., Novita AI, Together AI and SambaNova. At MiniMax, call it as MiniMax-M3.