Retrieve all served vLLM models and LoRA adapters from a specified inference target, with adapters flagged, for governance-grade AIops monitoring and root-cause analysis.
MIT
519,972 tools. Updated 2026-09-06 08:15
"Understanding Inference Models" matching MCP tools:
- Check Ollama server status and available models to see if free local inference is available for processing large contexts.MIT
- Verify connectivity to CLIProxyAPI and availability of configured companion models, without running inference.MIT
- Search AI inference models and get current per-token prices, filtering by name, tier, or price to compare providers.MIT
- List models available for inference via Tuning Engines API. Includes platform models and your deployed trained models.MIT
- List available Claude Code CLI models, including identifiers, resolved models, descriptions, and supported effort or mode capabilities, without running an inference turn.MIT
Matching MCP Servers
- AlicenseAqualityAmaintenanceIntegrates Xiaomi MiMo's multimodal API to enable understanding of images, audio, and video through natural language prompts.3MIT
- AlicenseAqualityCmaintenanceA minimal MCP server that exposes Gemini's web and multimodal understanding (video, audio, document, image) through five dedicated tools.51MIT
Matching MCP Connectors
x402 LLM proxy + data-enriched analysis (17 sources) + TimesFM predictive IoT intelligence.
Check if a task runs locally vs cloud. Save money on calls that don't need cloud inference.
- Send chat messages to run text inference using multiple LLMs. Choose from 11 models priced between $0.003-$0.10 USDC per request.MIT
- Retrieve a list of models currently loaded in memory and ready for inference.-
- Perform high-speed chat completions using Groq's inference with open models such as Llama 3, Mixtral, and Gemma.AGPL 3.0
- Search and use Hugging Face models, datasets, and spaces through inference, information retrieval, and authentication operations.Mozilla Public 2.0
- Estimate the USDC cost of an inference call before making it, using model and billing units (tokens, characters, or audio seconds), to inform gating decisions.MIT
- Find available AI and voice models on Floe Inference, including their IDs, modalities, and context windows, to use with OpenAI-compatible endpoints.MIT
- Break down an agent session's token usage per inference turn, returning per-turn entries and an aggregated summary while excluding non-inference events like tool calls.GPL 3.0
- Generates detailed textual descriptions of one or more images for text-only models, enabling scene/UI understanding, OCR, comparison, and structured extraction from multi-image uploads.MIT
- Retrieve daily LLM usage, cost trend, and model breakdown for a specific agent to optimize inference spending.MIT
- Identify running models and engine server details across vLLM, SGLang, and TGI. Provides engine-agnostic access to model IDs and server information from /v1/models or /info.MIT
- Retrieve real-time health metrics for inference providers: health score, breaker state, uptime ratio, and recent failures.MIT
- Load a model into memory for inference, with optional keep-alive duration and provider selection.Creative Commons Attribution Non Commercial No Derivatives 4.0 International
- List, load, or unload models on a local LLM server to switch active models in LM Studio without opening the GUI.Apache 2.0
- List all AI models supported by the AiChat API, grouped by provider, so you can see which models are available for your conversations.MIT
- Retrieves KV-cache utilization, prefix-cache hit rate, and preemption count to monitor inference cluster health and diagnose performance issues.MIT