MiniMax M3
MiniMax's fast-tier model: $0.30 per million input tokens and $1.2 per million output tokens, with a 1M-token context window.
Input / 1M tokens$0.30
Cached input / 1M$0.06
Output / 1M tokens$1.2
Example task$0.42$0.23 with 80% cached
| Maker | MiniMax · official pricing |
|---|---|
| Context window | 1M tokens · up to 128K output |
| Capabilities | tool calling · reasoning · images |
Prices from LiteLLM's open price list, updated Oct 9, 2026. "Example task": 1M input + 100K output tokens. Check the provider's page before you budget.
Agents that take any provider
These work with OpenRouter or any OpenAI-compatible endpoint, so they can usually run MiniMax M3 too.
OpenAI Codex LangChain Pi OpenCode GitHub Copilot LangGraph AI SDK Zed Mastra Deep Agents CopilotKit Cloudflare Agents SDK Oh My Pi DSPy OpenAI Agents SDK Junie Agno Cline CrewAI mini-SWE-agent OpenHands eve LlamaIndex Agent Development Kit (ADK)
Cheaper fast-tier alternatives
Claude Haiku 5.5 · $0.15 GPT-6 Luna · $0.15 MiMo V2.6 Flash · $0.17 GLM-5.3 Flash · $0.20