Skip to main content
Glama
711,321 tools. Updated 2026-10-28 06:09

"wgpu" matching MCP tools:

  • neon3_nui_submitC

    neon3-mcpB

    Validate NUI Flow locally, then submit it through ui-runtime to the WGPU runtime for deployment.
    MIT

  • get_pricesA

    GPU Finder MCP serverA

    List all provider offers for a GPU model and node size, with instance type, region, GPU count, on-demand/spot USD price per node hour, links, and update time for full price comparison.
    MIT

  • get_data_statusA

    GPU Finder MCP serverA

    Check how fresh GPU Finder pricing and availability data is, including each provider's last success, consecutive failures, and last error, to decide whether an answer may be stale.
    MIT

  • get_availabilityA

    GPU Finder MCP serverA

    Tracks daily GPU stock availability over the past seven days by provider and node size, showing observed available versus total instances to help judge how reliably a GPU can be found.
    MIT

  • gpu_submitA

    gpu-routerA

    Submit and run a script on a free cloud GPU for training, fine-tuning, or evals; it packages the project, installs dependencies remotely, and keeps the job running.
    MIT

  • gpu_statusA

    gpu-routerA

    Check the state of a GPU training job by id, or get an overview of active and recently finished jobs plus provider health. Returns progress, events, and next-step guidance.
    MIT

Matching MCP Servers

  • wgpu-mcp

    Ping-2o
    A
    license
    Not graded
    quality
    C
    maintenance
    Provides AI agents with live access to wgpu crate documentation from docs.rs, enabling search, item and member lookup, listing, and version/info queries through MCP tools.
    MIT

  • mcp-colab-gpu

    mio-github
    A
    license
    A
    quality
    D
    maintenance
    Enables MCP-compatible AI assistants to run Python code on Google Colab GPU/TPU runtimes, supporting accelerators like T4, A100, H100, with background execution and Google Drive integration.
    10
    28 PyPI
    3
    MIT

Matching MCP Connectors

  • GPU Finder

    Live cloud GPU prices, observed stock and 12-month price history across 30+ providers (H100, H200, A100, RTX 5090 and more), refreshed hourly. Read-only. Needs a GPU Finder API key sent as "Authorization: Bearer <key>"; free key with 60 data calls a month at https://gpufinder.dev/developers. Listing tools and get_usage are free.

  • Lium GPU Cloud: live GPU prices and rent links

    Live prices and rent links for H100, H200, B200 and RTX 4090 GPUs on Lium, plus Lium docs search.

  • gpu_fetchA

    gpu-routerA

    Ensure a finished GPU job's outputs are on this Mac and list them, downloading again only when missing or requested; unfinished jobs show where outputs will land.
    MIT

  • gpu_quotaA

    gpu-routerA

    Check free GPU quota per provider: used, limit, unit, reset time, and live vs estimated, to plan long jobs and answer users before routing.
    MIT

  • gpu_inferA

    gpu-routerA

    Route a single LLM chat completion to the free inference provider with the most daily quota, falling back automatically on rate limits or outages.
    MIT

  • gpu_cancelA

    gpu-routerA

    Stops or cancels a running GPU job to stop free-tier quota use; finished jobs are unchanged and existing outputs/checkpoints are kept.
    MIT
    Destructive

  • gpu_logsA

    gpu-routerA

    Fetch recent stdout and stderr from a GPU job, bounded to avoid context floods. Use the since cursor to follow a running job and tail to limit lines.
    MIT

  • gpu_routeA

    gpu-routerA

    Preview where a training job would run and why, without submitting it. Shows routing candidates, approval rules, and which files would ship.
    MIT

  • comfy_batchA

    GPU-Gate-MCPA

    Run ComfyUI jobs on a shared GPU by unloading the local LLM, blocking until all jobs finish, and returning per-job status and output file paths.
    MIT

  • list_workersA

    agent-gpu-poolB

    List available GPU workers and their normalized capabilities, excluding credential profiles, to match hardware for job submission.
    MIT

  • list_workersA

    agent-gpu-poolB

    Retrieve normalized capabilities of available GPU workers to support selection of eligible workers for job submission, without requiring credential profiles.
    MIT

  • voice_batchA

    GPU-Gate-MCPA

    Run VoiceStudio speech jobs on a shared GPU by unloading the local LLM, then block until all finish and return per-job status plus output paths.
    MIT

  • gate_statusA

    GPU-Gate-MCPA

    Check read-only GPU gate state and running batch progress before starting jobs, including LLM load status, ComfyUI/VoiceStudio availability, per-process GPU memory, warnings, and last batch result.
    MIT

  • sync_quotasB

    agent-gpu-poolB

    Refresh provider quotas to ensure scheduling decisions use authoritative limits, with unknown values treated as zero.
    MIT

  • cancel_jobB

    agent-gpu-poolB

    Request cancellation of a GPU job. The job status updates only after the worker confirms the cancellation.
    MIT

  • sync_quotasB

    agent-gpu-poolB

    Refresh authoritative provider quotas to ensure accurate GPU resource availability, treating missing data as zero to prevent overcommitment.
    MIT