MiniMax: MiniMax M2.7
minimax/minimax-m2.7 — available through the Autonomous Relay Agents API, served by MiniMax.
MiniMax: MiniMax M2.7 has a 204,800-token context window and costs $0.30 per million input tokens and $1.20 per million output tokens, at the provider's list price on your own key.
- Context length
- 204,800 tokens
- Input
- $0.30 per M
- Output
- $1.20 per M
- Served by
- minimax
- Input modalities
- text
Variants
The same model under another model id, priced or limited differently. Call it by its id; prices are per million tokens.
| Variant | Model id | Context | Input | Output |
|---|---|---|---|---|
| MiniMax: MiniMax M2.7 (free) | minimax/minimax-m2.7:free | 196,608 | $0.00 per M | $0.00 per M |
This is a BYOK-only model: requests use your own provider key. Prices shown are the provider's list prices.
Call it
Any OpenAI-compatible client works: change the base URL and send this model id.
curl https://autonomousrelay.com/v1/chat/completions \
-H "Authorization: Bearer $AUTONOMOUS_RELAY_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "minimax/minimax-m2.7", "messages": [{"role": "user", "content": "Hello"}]}'Compare
- MiniMaxAI/MiniMax-M2.7-Turbo vs MiniMax: MiniMax M2.7
- Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo vs MiniMax: MiniMax M2.7
- deepseek-ai/DeepSeek-V3 vs MiniMax: MiniMax M2.7
- MiniMax: MiniMax M2.7 vs Qwen: Qwen3.5-122B-A10B
- MiniMax: MiniMax M2.7 vs Qwen: Qwen3.5-35B-A3B
- MiniMax: MiniMax M2.7 vs Qwen: Qwen3.6 27B