MiniMax-M3
by MiniMaxChat
MiniMax multimodal model for long-context coding, perception, and agent planning
- Context
- 1M tokens
- Max output
- 512,000 tokens
- Parameters
- 427B total, 22B active
- Released
- 2026-06-01
- Weights
- Open
- Licence
- minimax-community
minimax/minimax-m3 · also seen as MiniMaxAI/MiniMax-M3, accounts/fireworks/models/minimax-m3, accounts/fireworks/routers/minimax-latest, minimax-m3@eu, +8 more
Specifications
Full facts as recorded, incl. what is unknown.
- Released
- 2026-06-01
- Weights
- Opennot gated
- License
- minimax-community
- Input
- Text, Image, Video
- Output
- Text
- Context window
- 1,000,000 max across endpoints
- Max output
- 512,000
- Providers
- 41 + 25 gateways, not counted as providers
- Epoch Capabilities Index
- 147.1 ± 3.6 against price
- Direct price range
- $0.20–$0.60 in · $0.90–$2.40 out per 1M tokens · 36 of 41 providers have a comparable price
More details
- Last updated
- 2026-09-15
- Knowledge cutoff
- 2025-01
- Weights link
- Hugging Face · MiniMaxAI/MiniMax-M3
- Parameters
- 427B total, 22B active
- Developer country
- China
Capabilities
Across all endpoints: supported if any endpoint supports it.
- Tool calling
- Parallel tools
- Structured output
- Reasoning
- Image input
- PDF input
- Audio input
- Audio outputunknown
- Video input
- Prompt caching
- Web searchunknown
- Computer useunknown
- System messages
- Assistant prefillunknown
- Temperature
- Streaming
- Attachments
Endpoints
One row per provider and variant; gateways are listed separately.
Loading endpoints…
Price history
Atom feedNo change since 19 Oct 2026. The cheapest direct price has been $0.20 in / $0.90 out per 1M tokens at Vultr on every data refresh since then.
Tracked from 19 Oct 2026. Headline-eligible direct prices only, so a price the sanity check flags never draws a spike. Series as JSON
Cost calculator
Loading endpoints…
Similarly priced models
Blended rate per 1M tokens, each at its own cheapest provider (10M input : 2M output mix).
| Model | Blended/1M | In/1M | Out/1M | Context | vs MiniMax-M3 |
|---|---|---|---|---|---|
| MiniMax LatestMiniMax | $0.352 | $0.23 | $0.96 | 1M | +0% |
| MiniMax Text 01MiniMax | $0.35 | $0.20 | $1.10 | 1M | −0% cheaper |
| Step 3.7 FlashStepFun | $0.358 | $0.20 | $1.15 | 256K | +2% |
| Qwen3-VL PlusAlibaba (Qwen) | $0.343 | $0.137 | $1.37 | 262K | −3% cheaper |
| Llama 4 Maverick 17b 128e InstructMeta | $0.362 | $0.24 | $0.97 | 1.05M | +3% |
| Trinity Large ThinkingArcee AI | $0.342 | $0.25 | $0.80 | 262K | −3% cheaper |
| Llama 3.1 Nemotron 70B InstructNVIDIA | $0.366 | $0.357 | $0.408 | 128K | +4% |
| Qwen2.5 72B InstructAlibaba (Qwen) | $0.367 | $0.36 | $0.40 | 131K | +4% |
More from MiniMax
Newest MiniMax models in the catalogue. All MiniMax models
| Model | Released | Context | Listed price |
|---|---|---|---|
| MiniMax-M3.1-Flash | 2026-10-19 | 1M | from $0.15 |
| MiniMax-M3.1-Flash-Preview | 2026-09-27 | 1M | unpriced |
| H3 Max | 2026-09-02 | — | unpriced |
| MiniMax H3 Max | 2026-08-27 | — | unpriced |
| MiniMax H3 | 2026-07-31 | — | unpriced |
| H3 | 2026-07-29 | — | unpriced |
| Speech 2.8 Turbo | 2026-07-16 | — | unpriced |
| Speech 2.8 HD | 2026-07-16 | — | unpriced |