MiniMax M3.1 Flash
@sdn/minimax-m3.1-flashGeneral purposeLive in the catalogMiniMax released M3.1 Flash on 19 October 2026 as an open-weight model for everyday agent and coding work. It reads up to a million tokens, takes images, and can call several tools in one turn. It always thinks before it answers: reasoning_effort runs from low to max, and max is what you get when you send none. Repeated prompt prefixes are cached automatically, so a long agent session costs fewer credits from the second turn on.
- 01A million-token window for long sessions and big repos
- 02Several tool calls in one turn
- 03Thinking depth you set, from low to max
Strict, schema-faithful tool calls, enforced by the engine on every request — safe to build an agent loop on.
Thinks before it answers. Budget max_tokens generously — hidden reasoning counts against it.
Reads images — screenshots, diagrams, and UI states — inline in the conversation.
One endpoint. Served by our engine.
MiniMax M3.1 Flash is served through the Sideren engine — zero-downtime serving is the design target, not a status-page apology. You request it by name; everything else is our problem.
curl https://api.sideren.io/v1/messages \
-H "x-api-key: sdn_your_key" \
-H "content-type: application/json" \
-d '{
"model": "@sdn/minimax-m3.1-flash",
"max_tokens": 1024,
"messages": [{ "role": "user", "content": "Hello" }]
}'OpenAI-style clients work too — send the same model name to /v1/chat/completions with a Bearer key. See the docs for both dialects.
Your agent stays the same.
Its inference gets better.
Start free · no card · Paid plans from $5/mo