Choose theme
AI Visuals
MiniMax CLI
MiniMax CLI provides terminal commands and a TypeScript SDK for MiniMax’s hosted AI services.
It sends platform requests rather than downloading the underlying models for local inference. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
A client for hosted requests
The package supplies mmx commands and a TypeScript interface.
Why it stands out
Task and output handling
Documented commands cover text, image, video, speech, vision and search. Published examples distinguish video task IDs, transcript formats and different quota response shapes. The v1.0.21 release retired Music CLI and SDK support.
Availability
Node.js client with platform authentication
The README lists Node.js 18+ and MiniMax Token Plan access.
Why it matters
What makes it useful
For terminal authentication, the README provides mmx auth status. It says OAuth and saved API-key login are mutually exclusive: logging in with one clears the other. Choose your workflow’s login method rather than expecting both saved methods to remain active.
What to know
Where it fits
For a video job that continues in the background, the SDK’s async example returns a taskId immediately. A later getTask call checks that task; downloading uses a file ID. Its MiniMax-H3 example also passes the model when checking the task. Preserve that task context rather than treating the initial ID as a finished video.
Notable points
What stands out
For speaker-labelled transcription, the README specifies verbose_json plus --output json to display speakers and segment timestamps. srt and vtt return subtitle documents. The default json response prints transcript text, so choose the output matching your task.
Before using
What to review
When displaying SDK quota, check its response type. The SDK guide’s Token Plan and OAuth example reads model_remains; sk-api- credentials instead return an account balance with available_amount. Its supplied type guards distinguish the two shapes without changing the API response. For a service or CI job whose home directory differs from where you logged in, the README recommends setting MMX_CONFIG_DIR to the directory containing config.json. This selects where that process looks for the saved configuration. Check the account region, credential access, request costs and permitted automation before submitting prompts, files, media or search queries. Choose what submitted content and generated outputs the workflow retains.
Reader fit
Who may find it relevant
It fits terminal and TypeScript developers automating MiniMax's hosted text, media or search workflows. For scripts, the error reference separates exit code 2 for invalid arguments, 3 for authentication, 4 for quota, 5 for timeout and 6 for network errors. Use the documented code to identify which step failed rather than treating every nonzero exit as the same problem.
Editorial note
Why LifeHubber lists it
We list MiniMax CLI for media builders who already have a character image and a motion clip to include in a video request. Its H3 command example supplies --reference-image character.png and --reference-video motion.mp4 together with a prompt. This gives a scripted media workflow distinct reference inputs to carry into one generation job, beyond polling a task ID and downloading its result. Supplying those references does not establish that the generated video will preserve the character or reproduce the motion.
Source links
Source materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
More in Ecosystem
Keep browsing this category
Explore more AI ecosystem resources.
Laya
NandhaKishorM/laya
An early Apache-2.0 model family and Python runtime for bounded choice, score, and yes-or-no decisions, with English, multilingual, and task-specialized checkpoints plus a router that selects between them.
CLM
Contrastive-LM/CLM
An Apache-2.0 contrastive model and local server for typed decisions and candidate ranking, with separate state and action embeddings, reusable candidate caches, a browser playground, public reference heads, and fine-tuning tools.
OpenJev
razorback16/openjev
An early open-source server for typed, probability-based yes-or-no, choice, and score decisions, using DiffusionGemma behind a Jev-compatible API for routing, filtering, and scoring steps.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
