M2.5 HighSpeed for tight latency budgets.
- Tier
- flash
- relative cost band
- Context
- 200K
- tokens per request
- Included from
- Free
- free plan
- Credits
- 3,000/mo
- a month on Go
Plan access
Each plan grants real usage on a schedule set by the plan (weekly for every paid plan, monthly on Free). One credit is one hundredth of a dollar of real model usage. A call to MiniMax M2.5 HighSpeed costs whatever tokens it actually used.
| Plan | Included | Multiplier | Per 7 days | Credits / mo |
|---|---|---|---|---|
| Free$0/mo | Yes | — | — | 300 |
| Go$5/mo | Yes | 1× | 750 | 3,000 |
| Plus$25/mo | Yes | 5× | 3,750 | 15,000 |
| Pro$50/mo | Yes | 10× | 7,500 | 30,000 |
| Max$100/mo | Yes | 20× | 15,000 | 60,000 |
Call it
The endpoint speaks the OpenAI chat completions protocol, so any client you already have works once you change the base URL and the key.
curl https://ai.eaon.dev/v1/chat/completions \
-H "Authorization: Bearer $EAON_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "eaon/minimax-m2.5-highspeed",
"messages": [{"role": "user", "content": "Hello"}]
}'