Skip to main content
Glama
limam-B

mlagents-mcp

by limam-B
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to mlagents-mcp

No user-submitted related servers found.

    Related Servers

    • projtool

      zhyx12
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables ML researchers to manage experiments across local and remote AutoDL GPU instances, including experiment creation, training launch, run polling, and report writing via Claude Code.
      1
      MIT

    • ml-lab-mcp

      CyrusTao
      F
      license
      Not graded
      quality
      B
      maintenance
      Enables LLMs to manage and run machine learning training jobs on a remote server, including syncing code, submitting experiments, monitoring progress, reading TensorBoard metrics, and receiving completion notifications.
      -

    • gpuctl-mcp

      EricYuan2007
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI agents to manage GPU training end-to-end through natural language, including submitting and scheduling jobs, monitoring logs and metrics, diagnosing failures, comparing runs, and recommending the best checkpoints.
      Apache 2.0

    • claude-sessions-mcp

      LukeLamb
      A
      license
      A
      quality
      C
      maintenance
      Lets Claude manage tmux sessions on Linux — start long-running ML training jobs, check their output, send keystrokes, and kill them.
      7
      149 npm
      MIT
    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.9/5.0

    Scored across 18 tools

    Disambiguation4/5

    Most tools target distinct resources/actions: launching, stopping, resuming, config, waiting, checking, logging, comparing. A coupple of overlaps exist, especially check_completion vs. wait_for_completion and the various check_* tools vs. get_run_status, but descriptions clearly separate blocking/non-blocking and specific conditions.

    Naming Consistency5/5

    All tool names follow a consistent snake_case verb_noun pattern: force_training, stop_training, update_config, check_step, get_metrics, list_runs, export_model. There are no style mixes or vague one-word names.

    Tool Count3/5

    18 tools is in the heavier range and the count is inflated by several near-duplicate monitoring utilities: check_step, check_reward, check_completion, check_checkpoint, wait_for_completion, and wait_for_first_metrics could potentially be consolidated. Still, the coverage is understandable for a training lifecycle server.

    Completeness4/5

    The server covers most of the training lifecycle: launch, resume, stop, configure, monitor, wait, list, compare, and export models. Minor gaps exist, such as no explicit delete/cleanup for run artifacts besides process cleanup, but core workflows are not dead-end.

    Maintenance

    ActivityInactive
    ResponsivenessNo issues

    Resources