Setup guide - From MiniMax and Zed docs
Use MiniMax in Zed
MiniMax documents adding an OpenAI-type LLM provider in Zed with API URL https://api.minimax.io/v1; Zed documents the matching language_models.openai_compatible settings and key handling.
Last verified 5 steps3 official sources
Steps
- Step 1.
Install Zed per the official docs.
- Step 2.
Settings > LLM Provider > +Add Provider > choose OpenAI, then fill API URL, API Key, and Model Name.
SnippetAPI URL = https://api.minimax.io/v1 Model Name = MiniMax-M3.1-Flash-Preview - Step 3.
Click Save Provider. Then back in the LLM Provider list, click the new MiniMax entry, re-enter the API Key and press Enter to confirm.
- Step 4.
Return to the agent panel, click Select a Model at the bottom-right and pick MiniMax-M3.1-Flash-Preview.
- Step 5.
Optional settings.json equivalent, using Zed's documented openai_compatible format with MiniMax values. Zed reads the key from the keychain or from an env var named after the provider ID in upper snake case plus _API_KEY (provider ID minimax uses MINIMAX_API_KEY).
JSON{ "language_models": { "openai_compatible": { "minimax": { "api_url": "https://api.minimax.io/v1", "available_models": [ { "name": "MiniMax-M3.1-Flash-Preview", "display_name": "MiniMax-M3.1-Flash-Preview", "max_tokens": 1000000 } ] } } } }
Caveats
- UI steps are from MiniMax's M Plan docs and use a Subscription Key; MiniMax-M3.1-Flash-Preview is M Plan only for now, so pay-as-you-go keys should use MiniMax-M3 or M2.x ids.
- The settings.json snippet is a combination of Zed's documented openai_compatible schema and MiniMax's base URL / model id; Zed requires max_tokens to be the model's context window (MiniMax lists 1,000,000 for M3.1-Flash-Preview and M3, 204,800 for M2.x).
- Keys saved through Zed are stored in the system keychain; non-empty environment variables take precedence over keychain values. The env var name depends on the provider ID you choose.
More setup guides
Same tool, other providers
Related reading
- MiniMax H3: An Omni-Modal Video Model With Native Audio, 2K Output, and Open Weights ComingMiniMax H3 turns text, image, video and audio input into 2K video with native stereo sound at 0.80 CNY per second, with open weights promised to follow.
- MiniMax M2.5 for Developers: The Anthropic-Compatible Budget Frontier ModelMiniMax M2.5 hits 80.2% on SWE-bench Verified and plugs into the Anthropic SDK with two environment variables. Here is what you need to know before switching.
- Zed Just Made Parallel AI Agents a Native Editor PrimitiveZed Parallel Agents adds a native Threads Sidebar that runs multiple AI agents in one window, with per-thread agent choice and worktree isolation.
- Zed: The Open Source Agentic IDEZed is not another Electron-based editor. It's built from the ground up in Rust, which means real performance without the memory bloat that plagues other IDEs.
FAQ
How do I use MiniMax in Zed?
MiniMax documents adding an OpenAI-type LLM provider in Zed with API URL https://api.minimax.io/v1; Zed documents the matching language_models.openai_compatible settings and key handling.
Which MiniMax model ids work in Zed?
The official docs read for this guide list MiniMax-M3.1-Flash-Preview, MiniMax-M3, MiniMax-M2.7 and MiniMax-M2.5. Requests go to https://api.minimax.io/v1.
Is this an official MiniMax and Zed setup?
The steps draw on both MiniMax's and Zed's docs. It was last checked against 3 official pages on October 17, 2026.
Sources
Last verified against 3 official pages.
- platform.minimax.iohttps://platform.minimax.io/docs/m-plan/other-tools.md
- zed.devhttps://zed.dev/docs/ai/use-api-access
- platform.minimax.iohttps://platform.minimax.io/docs/api-reference/text-openai-api.md
Newsletter
New models, set up the day they ship
One email per week. New model releases, the coding tools that support them and how to wire them up.