MiniMax M3.1 Flash API: pricing, base URL and quick setup
Use MiniMax M3.1 Flash on A2Agent with model ID minimax-m3.1-flash. The catalog lists a 1M context window. Check token rates below and verify the features your app needs.
Specifications and pricing
| Model ID | minimax-m3.1-flash |
|---|---|
| Platform | MiniMax |
| Context window | 1M |
| Tagged for | chat, coding, vision |
| Discount | -50% |
| Input / 1M tokens | |
| Output / 1M tokens |
USD per 1M tokens. Final billing depends on your API key group and metered usage.
Cheapest way to use MiniMax M3.1 Flash
The official price is $0.150 input and $0.600 output per 1M tokens. On A2Agent it is $0.075 and $0.300. A month of 10M input and 2M output tokens costs $2.70 at the official price and $1.35 on A2Agent.
Quick setup
curl https://a2agent.me/v1/chat/completions \
-H "content-type: application/json" \
-H "authorization: Bearer sk-your-a2agent-key" \
-d '{
"model": "minimax-m3.1-flash",
"messages": [{"role": "user", "content": "Reply with OK"}]
}'Using it from a coding client
Use minimax-m3.1-flash as the model ID in your coding client. Check a short request first, then test tool calls and image input with your own API key before relying on them.
How it compares within its family
These comparisons use catalog context limits, tags and reference rates, not independent quality benchmarks. Rates can change. Check current pricing above and test your workload; a higher price or a larger context window does not establish better results.
M3.1 Flash lists at $0.15 per million input tokens and $0.60 output, exactly half of M3 at $0.30 and $1.20. Both have a 1M window. M3 carries the agent tag; M3.1 Flash carries coding and vision. The MiniMax models page (checked October 22, 2026) describes M3.1 Flash as a multimodal coding model with a 1M context window and tunable thinking depth. That page does not list the input types or tool calling, and none of these features has been tested through this gateway. Before moving an existing workflow off M3, run the same tasks on both and count retries as part of the cost.
Other models on this platform
Questions
Which model ID should I use?
Use minimax-m3.1-flash. The minimax-m3 ID selects the separate M3 entry.
What is the context window?
1M tokens, per the MiniMax models page (checked October 22, 2026). M3 has the same window.
How does it compare with M3 on price?
It lists at half: $0.15 per million input and $0.60 output, against $0.30 and $1.20 for M3.
Does it support image input and tool calling?
MiniMax describes the model as multimodal, which is where the vision tag comes from. The models page does not mention tool calling. Neither has been tested through this gateway, so check the features your client needs with your assigned API key.
Billed per token. Top up from $1, no top-up fee.