MiniMax · Model
MiniMax M3 pricing, context window and specs
MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens, and reads up to 524,288 tokens per request. By blended price it is the only current MiniMax model in this directory.
- Input / 1M
- $0.30
- Output / 1M
- $1.20
- Cached / 1M
- $0.06
- Context
- 524,288
- Max output
- Not published
- Released
- 2026-05-31
Estimate your cost with MiniMax M3Compare all MiniMax modelsCount tokens for MiniMax M3
Pricing
MiniMax M3 API pricing
| Price per 1M tokens | Input | Output |
|---|---|---|
| Standard | $0.30 | $1.20 |
| Cached input (read) | $0.06 | – |
Output costs 4.0× the input price, so long replies drive the bill more than long prompts. If your requests share a long fixed prefix, such as a system prompt or tool definitions, caching cuts that part to 20% of the normal input price.
Examples
What MiniMax M3 costs in practice
| Workload | Tokens in / out | Requests/day | Per request | Per month |
|---|---|---|---|---|
| Support chatbot50% cached | 1,500 / 400 | 1,000 | $0.00075 | $22.81 |
| RAG app20% cached | 6,000 / 500 | 500 | $0.00211 | $32.12 |
| Coding agent80% cached | 40,000 / 2,000 | 200 | $0.00672 | $40.88 |
| Document summariser | 8,000 / 600 | 300 | $0.00312 | $28.47 |
Change any number in the cost calculator.
Context
Context window
MiniMax M3 accepts up to 524,288 tokens per request, roughly 393,216 English words or 786 pages. That budget is shared between your prompt and the reply, and no separate cap on the reply is published.
Filling the whole window with one prompt costs about $0.16 at list price, so for repeated long-document work, retrieval (sending only the relevant parts) or prompt caching is usually much cheaper.
Features
Capabilities
- Image input
- Yes
- Tool / function calling
- Yes
- Reasoning mode
- Yes
- Prompt caching
- Yes
- Structured output
- Yes
- Open weights
- Yes
- Accepts
- text, image, video
- Knowledge cutoff
- Not published
As declared by the provider’s API. It shows a feature exists, not how well it works.
API
MiniMax M3 model ID
- OpenRouter
minimax/minimax-m3- Open weights
- MiniMaxAI/Minimax-M3
Use the OpenRouter ID when calling it through OpenRouter; MiniMax’s own API may name it differently. See the model ID reference for the providers we document.
MiniMax
Where MiniMax M3 sits in MiniMax’s line-up
Ranked by blended price (three parts input to one part output, $0.525 per 1M for MiniMax M3), it is the only current MiniMax model in this directory. See every MiniMax model and price
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| MiniMax M3 (this page) | $0.30 | $1.20 | 524,288 | 2026-05-31 |
MiniMax’s current models with a page here, cheapest first. Older and superseded models are listed on the provider page and in the comparison table.
History
Earlier versions of MiniMax M3
The previous model in this line is MiniMax M2.7, first listed on 2026-03-18. Compared with it, MiniMax M3’s input price is 43% higher and its output price is 43% higher, and the context window went from 196,608 to 524,288 tokens.
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| MiniMax M2.7 | $0.21 | $0.84 | 196,608 | 2026-03-18 |
| MiniMax M2.5 | $0.27 | $1.08 | 200,000 | 2026-02-12 |
| MiniMax M2.1 | $0.30 | $1.20 | 204,800 | 2025-12-23 |
| MiniMax M2 | $0.30 | $1.20 | 196,608 | 2025-10-23 |
| MiniMax M1 | $0.55 | $2.20 | 1,000,000 | 2025-06-17 |
Earlier releases with the same name and size in our data, newest first. Dates are when our source first listed each model.
Save
Cheaper alternatives
- Gemma 4 26B A4BGoogle80% cheaper
- Qwen3.5-9BQwen79% cheaper
- Seed 1.6 FlashByteDance Seed75% cheaper
- GPT-6 LunaOpenAI62% cheaper
- Claude Haiku 5.5Anthropic62% cheaper
Current models from major providers, one per provider, that keep image input, tool calling, a reasoning mode, a usable context window and output limit. Compared by blended price.
Compare
Similarly priced models
- Muse Glimmer 30BMeta$0.30 / $1.20
- DeepSeek V4.1 FlashDeepSeek$0.30 / $1.20
- Qwen3.7 PlusQwen$0.32 / $1.28
- GLM 5.3 FlashXZ.ai$0.37 / $1.25
- Command A+Cohere$0.30 / $1.50
Current models from other major providers, closest blended price.
FAQ
MiniMax M3 questions
How much does MiniMax M3 cost?
MiniMax M3 costs $0.30 per million input tokens and $1.20 per million output tokens, with cached input at $0.06. A typical chatbot reply (1,500 tokens in, 400 out) costs about $0.00093. Use the LLM cost calculator for your own workload.
What is MiniMax M3’s context window?
524,288 tokens, about 393,216 English words or 786 printed pages, shared between your prompt and the reply. No separate limit on the reply is published, so the window is the limit.
Does MiniMax M3 support prompt caching and batch requests?
Yes, cached input is billed at $0.06 per million tokens (20% of the normal input price). No batch price is published for it.
What are cheaper alternatives to MiniMax M3?
With the same essentials (image input and tool calling), the cheapest options are Gemma 4 26B A4B (80% cheaper), Qwen3.5-9B (79% cheaper), Seed 1.6 Flash (75% cheaper). Cheaper doesn’t mean equivalent, so test them on your own prompts.
What is the API model ID for MiniMax M3?
On OpenRouter it is minimax/minimax-m3. MiniMax’s own API may use a different name, so check its model list before you deploy. Pass the ID as the model value in each request.
What did MiniMax M3 replace?
The previous MiniMax model in the same line is MiniMax M2.7, first listed on 2026-03-18. Compared with it, MiniMax M3’s input price is 43% higher and its output price is 43% higher. Prices are only part of the picture, so test the newer model on your own prompts before switching.
Is MiniMax M3 open source?
Yes, MiniMax M3 has open weights, so you can run it yourself or choose between several hosting providers. The prices here are a typical hosted price.