As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/minimax-m3/

# MiniMax-M3

MiniMax, MiniMax-M family. Status: current. Type: reasoning model. Released: 2026-06-01. Training cutoff: not published.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| MiniMax API (Anthropic format) | `MiniMax-M3` | https://api.minimax.io/anthropic | https://platform.minimax.io/docs/api-reference/text-anthropic-api |
| MiniMax API (OpenAI format) | `MiniMax-M3` | https://api.minimax.io/v1 | https://platform.minimax.io/docs/api-reference/text-openai-api |
| OpenRouter | `minimax/minimax-m3` |  | https://openrouter.ai/minimax/minimax-m3 |
| Hugging Face |  |  | https://huggingface.co/MiniMaxAI/MiniMax-M3 |

## Pricing

- Input: $0.30 per 1M tokens
- Output: $1.20 per 1M tokens
- Cache read: $0.06 per 1M tokens
- Unit as published: per 1M tokens (USD), standard tier, input <=512K (after permanent 50% discount); >512K input: 0.60/2.40/0.12. Priority tier 1.5x
- Source: https://platform.minimax.io/docs/guides/pricing-paygo
- Last checked: 2026-09-29

## Specs

- Input: text, image, video
- Output: text
- Context window: 1,000,000 tokens
- Max output: not published
- Open weights: yes
- Licence: minimax-community

## What it has not seen

- No published training cutoff. Counted from its release (2026-06-01): at least 1,826 AI events since (419 major, 82 historic), as of 2026-11-13.
- Everything since May 2026: https://postcutoff.com/since/2026-05/

## What stands out

- MiniMax Sparse Attention (MSA): New sparse attention for million-token contexts: 9x prefill and 15x decode speed-up vs M2 at 1M context, ~1/20 per-token compute. Source: https://huggingface.co/MiniMaxAI/MiniMax-M3
- Native multimodality from step one: Mixed text/image/video training from the start of pre-training (~428B total / ~23B active). Source: https://huggingface.co/MiniMaxAI/MiniMax-M3
- Three reasoning modes: thinking parameter selects among three reasoning modes; interleaved thinking with tool use. Source: https://huggingface.co/MiniMaxAI/MiniMax-M3

## Notes

MiniMax flagship LLM. OpenAI-format responses include <think> content that must be preserved across turns. MiniMax-M3.1-Flash-Preview (1M, tunable thinking) exists but only via Token Plan/MiniMax Code. Max output and knowledge cutoff not verified.

Low-cost 1M-context multimodal coding/agent model; Anthropic-SDK-first API, open weights.

```bash
curl https://api.minimax.io/anthropic/v1/messages \
 -H "x-api-key: $MINIMAX_API_KEY" -H "anthropic-version: 2023-06-01" -H "Content-Type: application/json" \
 -d '{"model":"MiniMax-M3","max_tokens":4096,"messages":[{"role":"user","content":"Hello"}]}'
```

Sources: https://platform.minimax.io/docs/guides/models-intro · https://platform.minimax.io/docs/guides/pricing-paygo · https://huggingface.co/MiniMaxAI/MiniMax-M3

Model page: https://postcutoff.com/m/minimax-m3/