MiniMax M3 pricing and routing profile
MiniMax M3 from MiniMax lists at $0.30 per million input tokens and $1.20 per million output tokens, with a 1,048,576-token context window. Prices from OpenRouter's catalogue on 2026-09-28.
Tool callingStructured outputReasoning controlLog probabilities
Cost per 1,000 requests
| Request shape | List cost |
|---|---|
| Classification or routing (600 tokens in, 10 out) | $0.192 |
| Chat reply (2,000 in, 400 out) | $1.08 |
| Retrieval-augmented answer (8,000 in, 500 out) | $3.00 |
Before prompt caching. Try your own shape in the cost calculator.
Can Gimbix route to it first?
Yes. 4 of 13 tracked hosts return token log probabilities for MiniMax M3, which is the confidence signal Gimbix uses to let a cheaper model answer first.
Measured by Gimbix
On 400 real BANKING77 customer messages, MiniMax M3 answered 316 correctly (95% interval 74.7%–82.7%), with 1 invalid answers, at $0.14 per 1,000 requests and a median of 1,529 ms. See all 22 models.
Who serves MiniMax M3
| Host | Input / output per 1M | Log probabilities | Uptime, last 30 min |
|---|---|---|---|
| CoreWeave | $0.23 / $0.96 | yes | 99.8% |
| GMICloud | $0.24 / $0.96 | no | 99.7% |
| DeepInfra | $0.28 / $1.10 | no | 97.1% |
| StreamLake | $0.30 / $1.20 | no | 99.7% |
| Venice | $0.30 / $1.20 | no | 86.6% |
| Together | $0.30 / $1.20 | no | 100.0% |
| Parasail | $0.30 / $1.20 | yes | 99.7% |
| AtlasCloud | $0.30 / $1.20 | no | 91.8% |
| Novita | $0.30 / $1.20 | no | 99.8% |
| Minimax | $0.30 / $1.20 | no | 99.9% |
| Mara | $0.60 / $2.40 | yes | — |
| SambaNova | $0.60 / $2.40 | yes | 96.9% |
| ModelRun | $0.75 / $3.00 | no | 99.2% |
A host that lists log probabilities may still omit them; Gimbix checks each one before relying on it.
Cheaper models that can answer first
Models priced below MiniMax M3 that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.
| Model | Input / output per 1M | Cheaper on a chat reply |
|---|---|---|
| DeepSeek V3 0324 | $0.29 / $1.14 | 4% |
| MiniMax M2.5 | $0.27 / $1.08 | 10% |
| DeepSeek V3.1 | $0.25 / $0.95 | 19% |
| Step 3.7 Flash | $0.20 / $1.15 | 20% |
| Qwen Plus 0728 | $0.26 / $0.78 | 23% |
| Qwen-Plus | $0.26 / $0.78 | 23% |
Compare MiniMax M3
- MiniMax M3 vs Claude Opus 5.5
- MiniMax M3 vs Claude Sonnet 5
- MiniMax M3 vs Claude Fable 5.1
- MiniMax M3 vs Claude Opus 5
- MiniMax M3 vs GPT-6 Sol
- MiniMax M3 vs GPT-6 Astra
- MiniMax M3 vs GPT-5.6 Sol
- MiniMax M3 vs GPT-5.6 Terra
- MiniMax M3 vs GPT-5.5
- MiniMax M3 vs GPT-5.4
- MiniMax M3 vs Gemini 3.1 Pro Preview
- MiniMax M3 vs Gemini 3.8 Flash
- MiniMax M3 vs Grok 4.7
- MiniMax M3 vs Qwen3.8 Max (0902)
- MiniMax M3 vs Kimi K3
- MiniMax M3 vs GLM 5.3
- MiniMax M3 vs DeepSeek V4 Pro 0423
- MiniMax M3 vs Mistral Large 3 2512
- MiniMax M3 vs Nova Premier 1.0
- MiniMax M3 vs Command A
- MiniMax M3 vs GPT-6 Luna
- MiniMax M3 vs GPT-5.6 Luna
- MiniMax M3 vs GPT-5.4 Mini
- MiniMax M3 vs GPT-5.4 Nano
- MiniMax M3 vs GPT-4.1 Mini
- MiniMax M3 vs gpt-oss-120b
- MiniMax M3 vs gpt-oss-20b
- MiniMax M3 vs Claude Haiku 4.5
- MiniMax M3 vs Gemini 3.5 Flash Lite
- MiniMax M3 vs Gemma 4 31B
- MiniMax M3 vs DeepSeek V4.1 Flash
- MiniMax M3 vs DeepSeek V4 Flash 0731
- MiniMax M3 vs GLM 5.3 Flash
- MiniMax M3 vs Hy4 preview
- MiniMax M3 vs MiMo-V2.6-Flash
- MiniMax M3 vs Nemotron 3.5 Lightning
- MiniMax M3 vs Nemotron 3 Ultra
- MiniMax M3 vs Qwen3.8 Flash
- MiniMax M3 vs Qwen3.7 Flash
- MiniMax M3 vs Qwen3.8 27B
- MiniMax M3 vs Kimi K2.6
- MiniMax M3 vs Mistral Small 4
- MiniMax M3 vs Ministral 3 8B 2512
- MiniMax M3 vs Llama 4 Maverick
- MiniMax M3 vs Llama 3.3 70B Instruct
- MiniMax M3 vs Nova Lite 1.0
- MiniMax M3 vs Nova Micro 1.0
- MiniMax M3 vs Command R7B (12-2024)
- MiniMax M3 vs Phi 4
Questions
How much does MiniMax M3 cost?
MiniMax M3 lists at $0.30 per million input tokens and $1.20 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $1.08 per 1,000 requests.
What is the context window of MiniMax M3?
1,048,576 tokens.
Can a cheaper model replace MiniMax M3?
Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match MiniMax M3 on one half of your requests and make at most three answers worse on the other.