{"id":51040065,"url":"https://github.com/graphsignal/graphsignal-profiler","last_synced_at":"2026-07-10T11:00:49.902Z","repository":{"id":41962075,"uuid":"357616470","full_name":"graphsignal/graphsignal-profiler","owner":"graphsignal","description":"Graphsignal Profiler","archived":false,"fork":false,"pushed_at":"2026-06-21T08:12:17.000Z","size":542,"stargazers_count":205,"open_issues_count":0,"forks_count":11,"subscribers_count":2,"default_branch":"main","last_synced_at":"2026-06-21T10:09:03.629Z","etag":null,"topics":["ai-agents","artificial-intelligence","debugging","deep-learning","huggingface","inference","langchain","langchain-python","machine-learning","monitoring","observability","openai-api","python","pytorch","tracer"],"latest_commit_sha":null,"homepage":"https://graphsignal.com","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/graphsignal.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":"NOTICE","maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2021-04-13T16:16:37.000Z","updated_at":"2026-06-21T08:12:22.000Z","dependencies_parsed_at":"2023-12-15T14:05:43.284Z","dependency_job_id":"b30efc0b-e755-4ab5-a8de-51171078ec51","html_url":"https://github.com/graphsignal/graphsignal-profiler","commit_stats":{"total_commits":179,"total_committers":2,"mean_commits":89.5,"dds":0.05027932960893855,"last_synced_commit":"d008cffa2c6d886e4b321c1844147f5d7cefe209"},"previous_names":["graphsignal/graphsignal","graphsignal/graphsignal-profiler"],"tags_count":177,"template":false,"template_full_name":null,"purl":"pkg:github/graphsignal/graphsignal-profiler","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/graphsignal%2Fgraphsignal-profiler","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/graphsignal%2Fgraphsignal-profiler/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/graphsignal%2Fgraphsignal-profiler/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/graphsignal%2Fgraphsignal-profiler/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/graphsignal","download_url":"https://codeload.github.com/graphsignal/graphsignal-profiler/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/graphsignal%2Fgraphsignal-profiler/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35329609,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-10T02:00:06.465Z","response_time":60,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agents","artificial-intelligence","debugging","deep-learning","huggingface","inference","langchain","langchain-python","machine-learning","monitoring","observability","openai-api","python","pytorch","tracer"],"created_at":"2026-06-22T10:00:28.895Z","updated_at":"2026-07-10T11:00:49.895Z","avatar_url":"https://github.com/graphsignal.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# Graphsignal: Inference Profiler\n\n[![License](http://img.shields.io/github/license/graphsignal/graphsignal-profiler)](https://github.com/graphsignal/graphsignal-profiler/blob/main/LICENSE)\n[![Version](https://img.shields.io/github/v/tag/graphsignal/graphsignal-profiler?label=version)](https://github.com/graphsignal/graphsignal-profiler)\n\n\nGraphsignal is a production-scale inference profiling platform that helps engineers optimize AI performance across models, engines, GPUs, and other accelerators. It provides essential visibility across the inference stack, including:\n\n* Continuous, high-resolution profiling timelines exposing operation durations and resource utilization across inference workloads.\n* LLM generation tracing with per-step timing, token throughput, and latency breakdowns for major inference frameworks.\n* System-level metrics for inference engines and hardware (CPU, GPU, accelerators).\n* Error monitoring for device-level failures and inference errors.\n* Inference telemetry for AI agents to identify bottlenecks and drive targeted improvements across the inference stack.\n\n[![Dashboards](https://graphsignal.com/external/screenshot-dashboard.png?v=2)](https://graphsignal.com/)\n\nLearn more at [graphsignal.com](https://graphsignal.com).\n\n\n## Install\n\n```bash\nUV_TOOL_BIN_DIR=/usr/local/bin uv tool install 'graphsignal[cu12]'   # CUDA 12.x\n# or\nUV_TOOL_BIN_DIR=/usr/local/bin uv tool install 'graphsignal[cu13]'   # CUDA 13.x\n```\n\n### Alternative: install into your workload environment\n\nIf you prefer a single environment, or you use the `graphsignal.watch()` Python API (which requires `graphsignal` importable by your application), install it directly into your workload's environment instead:\n\n```bash\npip install 'graphsignal[cu12]'   # CUDA 12.x\n# or\npip install 'graphsignal[cu13]'   # CUDA 13.x\n```\n\n\n## Profile\n\nWrap your launch command with `graphsignal-run`:\n\n```bash\nexport GRAPHSIGNAL_API_KEY=\u003cmy-api-key\u003e\ngraphsignal-run vllm serve \u003cmodel\u003e --port 8001\n```\n\nEnvironment variables read by the profiler:\n\n| Variable                              | Purpose                                                                |\n| ------------------------------------- | ---------------------------------------------------------------------- |\n| `GRAPHSIGNAL_API_KEY` *(required)*    | Your account API key.                                                  |\n| `GRAPHSIGNAL_TAG_\u003cKEY\u003e=\u003cvalue\u003e`       | Arbitrary tag attached to all signals (e.g. `GRAPHSIGNAL_TAG_DEPLOYMENT=us-prod`). |\n\nSign up for a free account at [graphsignal.com](https://graphsignal.com); you'll find the API key in [Settings / API Keys](https://app.graphsignal.com/settings/api-keys).\n\nSee the [Profiler CLI](https://graphsignal.com/docs/reference/profile-cli/) reference for the full set of options.\n\nApplications that bootstrap themselves can call `graphsignal.watch()` from Python instead — see the [Profiler API](https://graphsignal.com/docs/reference/profiler-api/) reference.\n\nSee integration documentation for libraries and inference engines:\n\n* [PyTorch](https://graphsignal.com/docs/integrations/pytorch/)\n* [vLLM](https://graphsignal.com/docs/integrations/vllm/)\n* [SGLang](https://graphsignal.com/docs/integrations/sglang/)\n\n\n## Optimize\n\n[Log in](https://app.graphsignal.com/) to Graphsignal to monitor and analyze your application.\n\n### Optimize with AI\n\nInstall the Graphsignal skill to let your AI coding agent (Claude Code, Codex, or Gemini) fetch and analyze signal context directly from your agent. See [AI Optimization](https://graphsignal.com/docs/guides/ai-optimization/) for setup instructions.\n\n\n## Overhead\n\nThe profiler has minimal impact on production performance. CUDA kernel activity is collected via CUPTI with low-overhead APIs, and analysis and upload happen in the sidecar process.\n\n\n## Security and Privacy\n\nThe profiler only establishes outbound connections to `api.graphsignal.com` to send data; inbound connections or commands are not possible.\n\nContent and sensitive information, such as prompts and completions, are not recorded.\n\n\n## Troubleshooting\n\nIf something doesn't look right, report it to our support team via your account.\n\nIn case of connection issues, please make sure outgoing connections to `https://api.graphsignal.com` are allowed.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgraphsignal%2Fgraphsignal-profiler","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgraphsignal%2Fgraphsignal-profiler","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgraphsignal%2Fgraphsignal-profiler/lists"}