Agent Skills and an MCP server for GPU performance profiling, benchmarking, optimization, and reporting, with an inference focus.

Local GPU translation Python library + MCP server — TranslateGemma via Ollama, 57 languages, zero cloud dependency

Read and analyze NVIDIA Nsight Graphics GPU Trace captures with Rust, JSON CLI, and MCP

Give any coding agent free Tesla T4 x2 GPUs via Kaggle's REST API -- MCP server, CLI, and skill.

MCP server that exposes NVIDIA GPU metrics (utilization, VRAM, temperature) as read-only tools via NVML or nvidia-smi

Measure Wine and macOS programs' GPU use on Apple silicon with Metal System Trace, from the shell or an MCP client

Measure Wine and macOS programs' GPU use on Apple silicon with Metal System Trace, from the shell or an MCP client

Open, daily index of GPU rental prices (H100, H200, B200, MI300X, A100) across hyperscalers, neoclouds and marketplaces, built from primary sources. Dashboard, cost calculator, MCP server.

MCP-first control plane for supported ComfyUI workflows with cryptographic model identity, explicit approvals, durable evidence, and no silent downloads.

**Run GPU workloads, deploy AI models, access inference APIs, and execute AI agents using Badgr's MCP server.** Badgr is an AI compute and workload execution platform for developers, coding agents, and AI assistants. Run workloads on cloud GPUs or compatible customer-owned GPUs (BYO GPU), with cost controls, execution logs, and verified results. **What can Badgr do?** - **GPU Compute / AI Workloads:** Run Python scripts, machine learning jobs, batch inference, and supported ComfyUI workflows. - **AI Model Deployment:** Deploy and serve LLMs using compatible runtimes such as vLLM and SGLang. Get a verified, OpenAI-compatible inference endpoint. - **AI API / LLM Inference:** Make AI API requests with response validation, supported fallback routing, and bounded execution. - **AI Agent Execution:** Launch coding and testing agents, run tasks, check acceptance criteria, and retrieve execution receipts. **Key features:** Cloud GPU execution, BYO GPU support, model serving, AI agent jobs, workload verification, execution logs, cost limits, and deployment management. Connect Badgr to your MCP-compatible AI assistant to execute real workloads instead of just generating instructions. Get started: https://aibadgr.com **Requires a Badgr API key. GPU workloads and other paid execution may incur usage charges.**

Free daily GPU rental price index (H100, H200, A100, B200, B300, RTX PRO 6000) and AI token price index. Website, MCP connector for Claude and ChatGPT, CC BY 4.0 data.

Remote MCP server: cloud GPU prices with live stock from 22 providers (grabgpu.com/mcp)

Agent-friendly CLI, MCP server and SDK for managing AutoDL GPU instances — create, boot, SSH, run and auto-shutdown, all scriptable. 用命令行和 MCP 管理 AutoDL GPU 实例。

One terminal for the GPUs of all your ssh hosts, with an MCP server so agents can find a free GPU. The remote side needs only nvidia-smi.

Remote MCP server that lets AI agents use the Meshive GPU Cloud

Kilawatt Cloud MCP server for requesting GPU compute capacity with account-based and x402 payment options.

Cooperative GPU queue for people and AI agents: one line per card, heartbeats, polite yields, MCP server. Home on Codeberg.

MCP server for ComputeNexa: let your AI assistant find, launch and stop private GPUs, with a price quote and your approval before every launch

MCP server that runs Python on Modal cloud hardware (CPU/RAM/GPU) — for Open WebUI and other MCP clients

Rent real GPUs from the terminal. Agent-first CLI + built-in MCP server for GPUsMarket — signup, rent, chat, pool. Per-second billing, $0.50 minimum per rental.

Rent real GPUs from the terminal. Agent-first CLI + built-in MCP server for GPUsMarket — signup, rent, chat, pool. Per-second billing, $0.50 minimum per rental.

Self-hosted homelab dashboard in one container — GPU, local-AI VRAM, Docker, systemd, energy costs. Read-only MCP server so AI agents can explore it.

Command-line interface and Python SDK for VoltageGPU: confidential GPUs on demand (Intel TDX, tenant-side NVIDIA attestation) and an OpenAI-compatible inference API, billed per second.

117 MCP tools for GPU infrastructure + Agent FinOps — deploy LLMs, manage VMs (Proxmox/XO/vSphere), track cost per agent, enforce budgets and model policies. Works with Claude, Cursor, n8n, LangChain

Give Claude Code, Codex and Gemini every machine you have: a live GPU, VRAM, RAM and disk inventory across your servers, SSH access between them, and an MCP server.

FIREFIGHTER: the coordinator and scheduler for unmanaged AI agents. Your AI agents won't set your GPU on fire. Or your CPU, your RAM, or your network. Claude Code, Codex, Gemini, Cursor, Kimi and more; MCP server; Windows/macOS/Linux.

MCP server for AmpleRun: AI agents find, rent and pay for GPUs (USDC on Base, x402), register their own account, and host their owner's GPU.

让 AI Agent 通过 MCP 自动租用 AutoDL GPU 实例、用完自动关机 | MCP server that lets AI agents lease AutoDL GPUs and power them off when done

Independent GPU rig grading: free preflight check, paid x402 grade, MCP server

MCP server for Forge GPU kernel optimization — generate and optimize Triton/CUDA kernels on real H100/A100 GPUs from any AI coding agent

Linux GPU shader with MCP control from LLMs

VibOps MCP server — control your GPU infrastructure from any MCP-compatible client

The sovereign, retained-mode, GPU-accelerated graphical user interface engine for Rust.

Remote MCP server for readygpu: live GPU, Mac and mini PC prices for running open AI models locally.

GPU-accelerated Linux desktops in Docker/WSL2 for AI agents and humans, featuring a built-in 60-tool MCP server and browser access.

No-auth remote MCP server for AI infrastructure data: LLM prices, GPU prices, AI funding. Built for AI from day one.

⁠Model Context Protocol (MCP) Server for Kilawatt Cloud—enables Cursor, Claude Desktop, and AI agents to programmatically request and manage GPU compute nodes.⁠

Zero-GPU-idle gateway for running ComfyUI on ephemeral Google Colab VMs with Zero-Trust Access

Semantic search for your vault and code, through the CLI or MCP. Inference runs on your GPU.

MCP server and CLI for VMware Private AI GPU/vGPU inventory, readiness, profiles, assignment and PAIS queries. Community-maintained.

Agent-friendly CLI, MCP server and SDK for managing AutoDL GPU instances — create, boot, SSH, run and auto-shutdown, all scriptable.

Train PyTorch on every GPU you can reach: one Trainer, from your laptop to many machines. NVIDIA CUDA, Apple Metal and CPU; self-hosted, open source.

Diagnose and Fix CUDA / GPU environments compatibility issues locally, in Docker, and CI/CD. CLI + MCP server available.

Claude Code and Codex plugin for deeplit: manage GPU inference instances, volumes, models and billing from your coding agent, authorized by a scoped API key.

MCP server (HTTP/SSE) for image generation with ComfyUI (SDXL) on your own GPU. Python + FastMCP. Self-hosted, MIT.

Denpex MCP Server: Diagnose GPU & ML training failures (CUDA OOM, NCCL timeouts, NVIDIA Xid, Slurm, vLLM) from Claude Desktop, Cursor, Windsurf, and Cline.

Search code and feature records by meaning through the command line or Model Context Protocol (MCP). Inference runs on your GPU.

Runtime evidence that helps agents trace, profile, and burn down hotspots in application and native code, GPU kernels, and inference stacks.

The first Task-Aware MCP server and automated VRAM calculator for LLM fine-tuning. Instantly snipe the cheapest, fastest GPUs across 10+ cloud providers.

Parallel video rendering with live dashboard, GPU auto-detection, checkpoint system, and stream-copy concat. Includes an MCP server, a Claude Code skill, and a CLI.

CPU and GPU stress test for Linux (GTK 4 / libadwaita) – live sensors and 30-minute graphs, power supply readings, safety limits, Vifte-aware, status icon and Claude (MCP) integration

Lead Green: the energy and CO2e cost of LLM tokens, per model, GPU and country. Open data, JSON API and MCP server behind leadgreen.ai. A Katonic project.

Self-tuning, disposable Docker sandboxes for AI agents. Runs code in locked-down containers sized to your hardware, with GPU sharing, crash cleanup, a CLI, a Python API and an MCP server for Claude.

MCP server that lets AI agents share one NVIDIA GPU — cross-session model claims, real busy detection via NVML, protected eviction, and full VRAM visibility for Ollama + everything else holding memory

GPU-accelerated local semantic code index for Claude Code. CUDA-powered embeddings, HNSW vector search, tree-sitter AST chunking, MCP server — sub-3ms queries, fully on your machine.

CPU-only LLM inference engine in pure Rust — 4-bit quantized models, hand-written AVX2 kernels, speculative decoding, and a local HTTP/MCP server. No GPU, no Python runtime.

Unofficial read-only MCP server for Zoomit (zoomit.ir/product, Zoobin): phone, laptop and GPU specs, shop prices in Toman, price history, comparisons. No API key. For Claude, Cursor, Copilot.

Make funky shader loops, posters, FastSD images and 9:16 product video ads with Claude, on your own computer. One MCP server + Claude Code plugin. No cloud, no accounts, no NVIDIA GPU.

Configurable tile-based system monitor for Windows: per-CPU, processes, GPU, kernel (DPC/ISR per driver, pool tags, objects), an hour of history, plugins and an MCP server for AI agents.

Provider-agnostic CLI and MCP server that signs, hash-chains, and verifies compute usage receipts (GPU-hours, workload type, hardware) so any third party can audit them without trusting the issuer.