MODELS / ABLITERATED
Minimax M3.1 Flash
Minimax M3.1 Flash, abliterated for cyber, red team, and agent testing. Input from $0.05 per 1M tokens. 1,048,576 token context.
minimax-m3.1-flash is Minimax M3.1 Flash with its refusal direction removed from the weights (abliterated), hosted for offensive cyber, AI red teaming, and agent testing. Same OpenAI-compatible endpoint and key as the rest of the lineup. The model id is the only change.
Key specifications
- Model id: minimax-m3.1-flash
- Display name: Minimax M3.1 Flash
- Input: $0.05 per 1M tokens
- Output: $0.25 per 1M tokens
- Cached input: $0.002 per 1M tokens
- Max context: 1,048,576
- Post-training: abliterated
- Endpoint: OpenAI-compatible chat completions
How to switch
Set model to minimax-m3.1-flash. Keep the base URL, the key, and your existing client. Guides for common libraries are at /docs.
CHAT COMPLETIONS
curl https://api.refuseless.com/v1/chat/completions \
-H "Authorization: Bearer $REFUSELESS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax-m3.1-flash",
"messages": [{"role": "user", "content": "Write a tight system prompt."}]
}'Compute is limited and first-come, first-served. Load credits early if you need sustained throughput.
Try the model that doesn't say no.
Free tier. OpenAI-compatible. Try Minimax M3.1 Flash in chat or the API.
