Pricing
| Provider | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|
| DeepInfra | $0.28 | $0.056 | $1.10 |
| Nebius | $0.30 | $0.06 | $1.20 |
| StreamLake | $0.60 | $0.12 | $2.40 |
USD, as of November 5, 2026. Live prices
MiniMax M3 API
- 524KContext
- 131KMax output
- $0.28/ $1.10From, per 1M in / out
- 3Providers
Providers
Cost
| Workload | DeepInfra |
|---|---|
| 1,000 chat replies2K in, 500 out each | $1.11 |
| 1,000 agent tasks50K in, 5K out each | $19.50 |
| Monthly batch job500M in, 50M out | $195.00 |
MiniMax M3 versions
| Model | Released | Context | Lowest input | Lowest output | Providers |
|---|---|---|---|---|---|
| MiniMax M3.1 Flashminimax/minimax-m3.1-flash | October 19, 2026 | 1M | $0.15 | $0.60 | 1 |
| MiniMax M3This pageminimax/minimax-m3 | June 1, 2026 | 524K | $0.28 | $1.10 | 3 |
Lowest listed price, USD per 1M tokens, as of November 5, 2026.
Quickstart
curl https://api.routerplus.com/v1/chat/completions \
-H "Authorization: Bearer $ROUTERPLUS_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax/minimax-m3",
"messages": [{ "role": "user", "content": "Hello" }]
}'Compare
FAQ
How much does the MiniMax M3 API cost?
On RouterPlus, MiniMax M3 costs $0.28 per million input tokens and $1.10 per million output tokens at DeepInfra, the lowest listed prices and the provider where routing starts. Prices as of November 5, 2026.
Which providers serve MiniMax M3?
3 providers serve MiniMax M3 on RouterPlus: DeepInfra, Nebius and StreamLake. By default the router tries them in routing order and fails over before the first byte, never mid-answer.
What is the MiniMax M3 model ID?
Use "minimax/minimax-m3" as the model with the OpenAI-compatible endpoint at https://api.routerplus.com/v1. The same key also accepts the Anthropic API format.
What is the MiniMax M3 context window?
524K tokens of context and up to 131K output tokens, per the RouterPlus catalog. A provider can list a lower limit; the console shows each one.
Is MiniMax M3 open weights?
Yes. MiniMax publishes MiniMax M3 as an open-weight model, which is why several independent providers can host it.
Do any MiniMax M3 providers retain prompts?
2 of 3 providers list no prompt retention (DeepInfra and Nebius). StreamLake may retain prompts. Pin the providers you accept with a routing policy.
Can I reserve capacity for MiniMax M3?
Yes. Tell us the model, the throughput you need and the region with "I need inference", and RouterPlus lines up reserved capacity for it.