Skip to main content
Glama
Sign In
metal-gpu-trace

metal-gpu-trace

by partprogramming09
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to metal-gpu-trace

No user-submitted related servers found.

    Related Servers

    • nsys-mcp

      KorovkoAlexander
      A
      license
      Not graded
      quality
      Not graded
      maintenance
      Enables GPU profiling and performance analysis via NVIDIA Nsight Systems, allowing agents to profile binaries and aggregate statistics for kernels, memory copies, and NVTX ranges. It supports advanced analysis through interval tree construction and structural queries on profiling reports.
      5
      MIT

    • Flameox

      Deboer6105
      F
      license
      Not graded
      quality
      A
      maintenance
      Enables coding agents to trace, profile, and investigate performance hotspots in application code, native code, GPU kernels, and inference stacks by coordinating local profilers, benchmark tools, trace processors, and direct capture targets. Returns bounded, typed runtime evidence with observed inputs, provenance, coverage, and limitations, all through explicit artifact paths or commands with no workspace setup required.
      -

    • hotkernel

      xinhuagu
      A
      license
      A
      quality
      B
      maintenance
      Enables coding agents to profile and disassemble GPU kernels, benchmark and compare candidate rewrites against baselines, and run an autonomous speedrun loop that verifies and records results.
      10
      32 PyPI
      Apache 2.0

    • InstrumentsMCP

      nemanjavlahovic
      A
      license
      A
      quality
      C
      maintenance
      Enables AI agents to profile iOS/macOS apps using Xcode Instruments, automate simulator interactions, and get structured performance data for optimization.
      37
      26 npm
      20
      MIT

    • io.github.morluto/flameox

      BrianDabrowski64376
      F
      license
      Not graded
      quality
      A
      maintenance
      Enables coding agents to turn explicit artifacts or commands into bounded local runtime evidence from profilers, benchmarks, trace processors, and direct targets, with typed provenance, coverage, and optional durable preservation. It exposes analysis and capture tools for tracing, profiling, and investigating hotspots in application and native code, GPU kernels, and inference stacks.
      -
    Deploy Server
    Shareable, User-Owned Context BackendPeople, applications, and agents read, write, and coordinate together in one context lake. Reach it over MCP, a CLI, a Python SDK, or REST.fulcradynamics.comAd

    TDQS

    A3.9/5.0

    Scored across 23 tools

    Disambiguation5/5

    Despite many perf_* tools, each has a clearly distinct role: record, attach, sweep, A/B compare, baseline, check, gate, history, diagnose, panel, export, and run listing. Similar-sounding pairs like perf_baseline versus perf_baseline_record are differentiated by whether they consume existing runs or record new ones. The trace and capture tools are also cleanly separated by operation and artifact type.

    Naming Consistency4/5

    The set follows a consistent snake_case convention with useful domain prefixes such as perf_, trace_, capture_, workload_, and gpu_. Names are not strictly verb_noun throughout—there are noun-style names like perf_runs, perf_history, and perf_panel—but the grouping is readable and predictable. Minor deviations do not impede tool selection.

    Tool Count4/5

    At 23 tools the surface is large, but this is a broad GPU profiling suite covering recording, comparison, baselines, gates, traces, captures, workload setup, and environment inspection. Each tool appears to earn its place, though the count is slightly heavy and could be streamlined. It is not an extreme mismatch for the stated scope.

    Completeness4/5

    The tool set covers the main lifecycle: fetching workloads, recording or attaching, saving baselines, comparing, checking, gating, diagnosing, visualizing, exporting traces, and inspecting raw trace/capture data. The main noticeable gap is management of saved artifacts, such as deleting or renaming recordings and baselines. That is a minor omission rather than a blocker for core profiling workflows.

    Maintenance

    ActivityMaintained
    ResponsivenessSyncing

    Resources