stealthstack.ai Status Publish a catalog

AI models and MLOps

Resources for working with AI models.

They cover inference gateways and routers, fine-tuning, local models, evals, prompt work, embeddings and RAG, and token and cost estimates.

Listings are what publishers declare in their ARD catalogs and the registries StealthStack reads. The category and summaries are our reading of each listing. Each listing links to its own page with the publisher's endpoint.

All types MCP serversSkillsA2A agentsAPIs
Inference gatewaysTraining and fine-tuningLocal modelsEvalsPrompt engineeringEmbeddings and RAGCost and tokens
  1. SKILLdatarobot-model-training Skill to train models and run AutoML experiments on DataRobot.
  2. MCP SERVERRoboflow MCP Server Hosted server for Roboflow: datasets, annotation, training, inference and Workflows.
  3. APICluster Protocol AI Gateway Pay-per-request inference gateway for 500+ models, payable in USDC via x402.
  4. SKILLhuggingface-llm-trainer Skill to fine-tune language and vision models with TRL or Unsloth on HF Jobs.
  5. SKILLhuggingface-vision-trainer Skill to fine-tune vision models for detection, classification, segmentation.
  6. APIRoboflow Serverless Inference OpenAPI OpenAPI spec for Roboflow serverless inference: models, Workflows, OCR, segmentation.
  7. SKILLPrompt Optimizer Skill to create and refine agent, system and developer prompts.
  8. SKILLhf-cli Skill for the hf CLI to manage models, datasets and Spaces on the Hub.
  9. SKILLtransformers-js Skill to run ML models in JavaScript or TypeScript with Transformers.js.
  10. MCP SERVERhuggingface/hf-mcp-server Access Hugging Face models, datasets, Spaces and papers.
  11. SKILLNeon Ai Gateway Skill for Neon AI Gateway: one API and credential for many LLM providers.
  12. APIVenice API OpenAI-compatible Venice API for text, image, video, audio; bearer key or x402.
  13. SKILLtrl-training Skill to train and fine-tune language models with TRL.
  14. MCP SERVEROpenRouter MCP server OpenRouter MCP: model catalog, provider uptime, rankings, benchmarks, docs, chat.
  15. SKILLhuggingface-local-models Skill to select models to run locally with llama.cpp and GGUF.
  16. SKILLhuggingface-community-evals Skill to run Hub model evaluations with inspect-ai and lighteval locally.
  17. SKILLtrain-sentence-transformers Skill to train sentence-transformers embedding, reranker and sparse models.
  18. SKILLtransformers Skill for Hugging Face Transformers inference and Trainer fine-tuning.
  19. SKILLazure-ai-openai-dotnet Skill for the Azure OpenAI SDK for .NET.
  20. SKILLazure-ai-ml-py Skill for the Azure Machine Learning SDK v2 for Python.
  21. SKILLnemo-automodel-distributed-training Skill configuring distributed training strategies in NeMo AutoModel.
  22. APIOpenRouter API OpenAPI description of the OpenRouter REST API: chat, models, providers, credits.
  23. SKILLagent-platform-eval-flywheel Skill for evaluating and improving models and agents on Google Cloud via an eval flywheel.
  24. SKILLdatarobot-feature-engineering Skill for feature engineering and feature discovery on DataRobot.
  25. SKILLeval-driven-dev Skill for evaluation-driven development of AI applications.
  26. SKILLdatarobot-model-explainability Skill to explain predictions and compute SHAP values on DataRobot.
  27. SKILLdatarobot-model-deployment Skill to deploy models and configure prediction environments on DataRobot.
  28. SKILLphoenix-evals Skill to build and run evaluators for LLM applications using Phoenix.
  29. SKILLLangsmith Evaluator Skill for building LangSmith evaluation pipelines.
  30. SKILLLangsmith Dataset Skill for creating and managing LangSmith evaluation datasets.
  31. SKILLAzure AI ML Python Skill for the Azure Machine Learning SDK v2 for Python: jobs, models, compute.
  32. SKILLhuggingface-trackio Skill to track and visualize ML training experiments with Trackio.
  33. SKILLdatarobot-predictions Skill to make predictions and batch score on DataRobot.
  34. API2s.io Ai 16 AI endpoints: multi-model chat, classify, OCR, image, PII; x402 USDC per call.
  35. SKILLjetson-inference-mem-tune Skill for picking LLM serving stacks and memory flags on NVIDIA Jetson.
  36. MCP SERVERVenice AI MCP server Venice MCP server for chat, image, video, audio, embeddings with x402 wallet auth.
  37. SKILLNemo Retriever Skill to search and extract across PDFs and scanned documents with NeMo Retriever.
  38. SKILLFind Models Skill to find AI models on Replicate.
  39. SKILLprompt-optimizer Skill turning rough ideas into finished prompts ready to send to an LLM.
  40. SKILLgoogle-cloud-solution-rag-enterprise-search-gke-sqldb Skill for designing a RAG enterprise search system on GKE with a vector SQL store.
  41. SKILLagent-platform-tuning Skill for fine-tuning Gemini or open models on Agent Platform.
  42. MCP SERVERTabArena MCP Server MCP server comparing machine learning models across tasks and datasets.
  43. SKILLOpen ASR Leaderboard Compares speech-recognition models across datasets and languages.
  44. SKILLazure-aigateway Skill for configuring Azure API Management as an AI gateway for models and MCP tools.
  45. SKILLjetson-speculative-decoding Skill for adding EAGLE-3 or draft-model speculative decoding to Jetson vLLM.
  46. SKILLVenice API skill Skill for calling the Venice API with model routing and x402 payment.
  47. SKILLjetson-llm-benchmark Skill for benchmarking LLM/VLM serving on Jetson across vLLM, llama.cpp and Ollama.

Related categories

Agent memory and contextAgent orchestrationAgent discoveryAgent trust and governanceAgent payments All categories