How to Use MiniMax M3 in Gemini CLI
MiniMax M3 suits repository debugging, performance tuning, and tasks that combine long text with visual evidence. Its million-token context and optional reasoning support work that spans code, documents, and repeated tool results.
Prompts to try with MiniMax M3 in Gemini CLI
Copy an example into Gemini CLI, or ask in your own words.
Use MiniMax M3 to investigate the failing tests in this repository. Trace the cause, make the smallest supported fix, and run the relevant tests.
Use MiniMax M3 to profile this project's slowest endpoint. Establish a baseline, optimize one measured bottleneck, and report correctness and timing results.
Use MiniMax M3 to implement pagination across this project's frontend and backend. Follow existing conventions and verify empty, first-page, and last-page behavior.
Use MiniMax M3 to compare the architecture documents with the implementation. Cite file paths for discrepancies and propose a prioritized correction plan.
Use MiniMax M3 with reasoning enabled to review this refactor. Trace changed call paths and identify behavior changes that existing tests do not cover.
How to set up MiniMax M3 in Gemini CLI
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the server
Or add it to ~/.gemini/settings.json yourself:
gemini mcp add --scope user --transport http spiral https://api.spiral.ai/mcp
- Sign in
Start gemini and run:
/mcp auth spiral
- Check it
In Gemini CLI, run /mcp. From your shell:
gemini mcp list
Use the Spiral skill
- Send Gemini CLI this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run MiniMax M3
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "minimax/minimax-m3", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Gemini CLI finds MiniMax M3 on Spiral, shows the price and runs it.
MiniMax M3 providers and prices
Spiral runs MiniMax M3 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.006404 per run at the default inputs. $0.39 per 1M input tokens, $1.56 per 1M output tokens.
Settings you can ask for
Ask Gemini CLI for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Min p | 0 to 1 | ||
| Top k | a whole number from 0 | ||
| Top p | 0 to 1 | ||
| Logprobs | on or off | ||
| Max tokens | 1 to 512000 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Top logprobs | 0 to 20 | ||
| Presence penalty | -2 to 2 | ||
| Frequency penalty | -2 to 2 | ||
| Repetition penalty | 0 to 2 |
Every API field of MiniMax: MiniMax M3 is on its tool page: open its card above.
What can you do with MiniMax M3 in Gemini CLI?
MiniMax M3 can carry a repository investigation from failing behavior through a tested change. Give it project context, measurable acceptance criteria, and feedback from tests or benchmarks.
Trace a bug across a repository
Ask MiniMax M3 to connect a failing test with the relevant implementation, propose a focused fix, and verify the behavior.
Optimize a measured bottleneck
Give MiniMax M3 profiling results and a correctness check. Ask it to change one bottleneck at a time and compare measurements.
Build a frontend and backend feature
Use MiniMax M3 to carry an acceptance checklist through interface changes, server logic, and tests.
Compare long documents
Ask MiniMax M3 to compare requirements across supplied documents, identify contradictions, and cite the passages behind each finding.
Explain charts in a research paper
Give MiniMax M3 chart images and the surrounding text. Ask it to separate reported results from interpretations and propose checks.
Review a software demonstration
Ask MiniMax M3 to inspect a demonstration video, describe the visible workflow, and identify confusing transitions for a usability review.
Prepare a sourced research brief
Have MiniMax M3 compare retrieved sources, reconcile disagreements, and draft a brief with evidence beside each conclusion.
More on MiniMax M3 and Gemini CLI
Overview
MiniMax M3 in other agents
Questions about MiniMax M3 in Gemini CLI
Can Gemini CLI use MiniMax M3?
Yes. Add Spiral to Gemini CLI as an MCP server, and Gemini CLI can find and run MiniMax M3 like any built-in tool.
How much does MiniMax M3 cost in Gemini CLI?
Up to $0.006404 per run at the default inputs. $0.39 per 1M input tokens, $1.56 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Gemini CLI plan is separate.
Do I need an OpenRouter account to use MiniMax M3 in Gemini CLI?
No. Spiral runs MiniMax M3 on its own OpenRouter account. You sign in to Spiral once and never paste a key.
How long does it take to set up MiniMax M3 in Gemini CLI?
About two minutes: add the server, then sign in, then check it.
Does Gemini CLI wait while MiniMax M3 runs?
MiniMax M3 runs in the background. Gemini CLI starts the run, checks on it and gives you the text, json when it's ready.
What is MiniMax M3 best suited to?
Start with debugging, frontend and backend development, performance tuning, and research that requires several tool steps. MiniMax also reports strong results on search and office workflows.
How much context can MiniMax M3 handle?
MiniMax documents a context window of up to one million tokens. That makes large code and document collections a useful target, especially when the task requires comparing evidence across them.
Can MiniMax M3 understand images and video?
Yes. It can reason over text, images, and video, which supports tasks such as explaining charts or reviewing a recorded software demonstration.
When should I use reasoning?
Use reasoning for debugging or analysis with several dependent steps. MiniMax M3 also supports adaptive reasoning, which lets the model decide when extra reasoning is useful, and a disabled mode for direct answers.
How long can MiniMax M3's output be?
MiniMax documents a maximum generation budget of 524,288 tokens and recommends 131,072. Reasoning shares that budget with the final answer, so the full allowance is not necessarily visible text.
Start using MiniMax M3 in Gemini CLI
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Gemini CLI:
set up https://spiral.ai/SKILL.md


