OptionalDevOpsVersion 1.0.0

Run 150+ AI Apps via the inference.sh CLI in Hermes Agent

Run 150+ AI apps (image, video, LLM) via inference.sh CLI.

Written by Neura Market from the official Hermes Agent documentation for Inference Sh Cli. Commands, paths, and version numbers are reproduced from the source unchanged.

Read the official documentation

Neura Market infsh CLI Reference

The infsh CLI lets you run over 150 AI applications in the cloud without managing your own GPU or individual provider APIs. You can generate images, video, 3D models, audio, perform AI-powered search, and more, all from a single command line.

Prerequisites

Before using the CLI, ensure it is installed and authenticated. Check your current status with:

infsh me

If the CLI is not installed, run the following commands:

curl -fsSL https://cli.inference.sh | sh
infsh login

For full setup details, see references/authentication.md.

Searching for Applications

Never guess an app ID. App IDs change and new applications are added frequently. Always search the catalog first using infsh app list --search with a keyword:

infsh app list --search flux
infsh app list --search video
infsh app list --search image

For a complete list of available apps, refer to references/app-discovery.md.

Running an Application

Once you have the correct app ID, run the application with the --input flag containing a JSON string of parameters. Always include the --json flag to get structured output for machine parsing.

infsh app run <app-id> --input '{"prompt": "your prompt here"}' --json

The --input flag is required and takes a JSON string. Properly escape quotes inside the string. Local file paths provided in the input (e.g., for images or audio) are automatically uploaded.

Image Generation

Search for image generation apps, then run the desired one:

# Search for image apps
infsh app list --search image

# FLUX Dev with LoRA
infsh app run falai/flux-dev-lora --input '{"prompt": "sunset over mountains", "num_images": 1}' --json

# Gemini image generation
infsh app run google/gemini-2-5-flash-image --input '{"prompt": "futuristic city", "num_images": 1}' --json

# Seedream (ByteDance)
infsh app run bytedance/seedream-5-lite --input '{"prompt": "nature scene"}' --json

# Grok Imagine (xAI)
infsh app run xai/grok-imagine-image --input '{"prompt": "abstract art"}' --json

Video Generation

Video generation can take 30 to 120 seconds. Warn the user about the wait before running these commands.

# Search for video apps
infsh app list --search video

# Veo 3.1 (Google)
infsh app run google/veo-3-1-fast --input '{"prompt": "drone shot of coastline"}' --json

# Seedance (ByteDance)
infsh app run bytedance/seedance-1-5-pro --input '{"prompt": "dancing figure", "resolution": "1080p"}' --json

# Wan 2.5
infsh app run falai/wan-2-5 --input '{"prompt": "person walking through city"}' --json

Local File Uploads

You can provide local file paths in the --input JSON. The CLI automatically uploads them. This works for upscaling, image-to-video, and avatar generation with audio.

# Upscale a local image
infsh app run falai/topaz-image-upscaler --input '{"image": "/path/to/photo.jpg", "upscale_factor": 2}' --json

# Image-to-video from local file
infsh app run falai/wan-2-5-i2v --input '{"image": "/path/to/image.png", "prompt": "make it move"}' --json

# Avatar with audio
infsh app run bytedance/omnihuman-1-5 --input '{"audio": "/path/to/audio.mp3", "image": "/path/to/face.jpg"}' --json

AI-Powered Search

Use Tavily or Exa for research and search tasks:

infsh app list --search search
infsh app run tavily/tavily-search --input '{"query": "latest AI news"}' --json
infsh app run exa/exa-search --input '{"query": "machine learning papers"}' --json

Other Application Categories

Search for additional categories such as 3D generation, audio/TTS, and Twitter/X automation:

# 3D generation
infsh app list --search 3d

# Audio / TTS
infsh app list --search tts

# Twitter/X automation
infsh app list --search twitter

Parsing Output

Always use the --json flag. The raw output is difficult to parse. When presenting results to the user, extract media URLs from the JSON and display them inline with MEDIA:.

Failure Modes

  • Authentication errors: Run infsh login or verify that the INFSH_API_KEY environment variable is set.
  • Command fails: The CLI may not be installed or authenticated. Check with infsh me.
  • Incorrect app ID: Always search first using infsh app list --search. An incorrect ID leads to failure.

Additional References

  • references/app-discovery.md
  • references/running-apps.md
  • references/cli-reference.md

More DevOps skills