onboard-model

Community

End-to-end onboarding of new LLM kernels.

AuthorRockyeast
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Orchestrates end-to-end onboarding of new LLM kernels into FlashInfer-Bench by coordinating model discovery, kernel definition generation, workload collection, and PR submission.

Core Features & Use Cases

  • End-to-end orchestration of model onboarding, kernel tracking, and repository updates across FlashInfer-Bench and related tools.
  • Automated generation of kernel definitions and workload collection pipelines, with optional PR submission workflows.
  • Real-world use: onboarding a new model with unknown kernels and ensuring kernels are defined, tested, and integrated into the benchmark suite.

Quick Start

Initiate onboarding of a target model with /onboard-model --discover or /onboard-model --model-name <model> to run the full pipeline.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: onboard-model
Download link: https://github.com/Rockyeast/flashinfer-trace-dev/archive/main.zip#onboard-model

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.