Back to Models
MiniMax M3
minimaxminimax/minimax-m3
MiniMax's 2026-06 frontier model. 1M context (min 512K SLA), native text + image + video input. Architecture reduces compute ~20× vs prior gen. SWE-Bench Pro 59.0%.
Context Length
1M
Max Output
66K
Input Priceper 1M tokens
$0.324/ 1M tokens
Output Priceper 1M tokens
$1.30/ 1M tokens
Modalities
textimagevideo→text
Capabilities
Vision
Pricing Breakdown
| Type | Rate |
|---|---|
| Prompts up to 512K tokens | |
| Input | $0.324 / 1M tokens |
| Output | $1.30 / 1M tokens |
| Cached input | $0.0648 / 1M tokens |
| Prompts over 512K tokens | |
| Input | $0.648 / 1M tokens |
| Output | $2.59 / 1M tokens |
| Cached input | $0.1296 / 1M tokens |
Tiered pricing: when a request's prompt (input, cached tokens included) is over 512K tokens, every token in that request is billed at the long-context rates; exactly 512K is still the base rate.
Supported Parameters
temperaturemax_tokenstop_ptoolstool_choiceresponse_formatstop
API Usage Examples
Use the global api.therouter.ai endpoint shown below for new integrations; the legacy China accelerated endpoint is retired.
cURL
curl https://api.therouter.ai/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer $THE_ROUTER_API_KEY" -d '{
"model": "minimax/minimax-m3",
"messages": [
{"role": "user", "content": "Summarize the key points from this input."}
]
}'