skills.rest
hao-ai-lab
hao-ai-labOFFICIAL
Explore all skills by this author →
Website𝕏@haoailabGitHub

add-model

Stage reference code, convert weights, and validate compatibility for FastVideo models.

3.9k|398|Updated Oct 24, 2024
Search Tags:#configuration#model integration#model porting#weight conversion#parity testing
One-click install
npx skills add https://github.com/hao-ai-lab/FastVideo --skill add-model-hao-ai-lab
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: add-model
Source: https://github.com/hao-ai-lab/FastVideo/tree/main/.agents/skills/add-model
Command: npx skills add https://github.com/hao-ai-lab/FastVideo --skill add-model-hao-ai-lab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) and scripts (resource) components.

What problem does it solve?

This Skill facilitates the process of porting new models or components into the FastVideo framework, ensuring compatibility, proper configuration, and seamless integration.

Core Features & Use Cases

  • Model Porting: Guides users through staging reference code and weights for new model families or components.
  • Conversion & Compatibility: Provides structured phases to handle weight conversion, configuration matching, and parity testing.
  • Use Case: For a researcher adding a novel video generation model, this Skill ensures the model ports correctly, passes parity checks, and integrates into FastVideo’s pipeline with minimal manual guesswork.

Quick Start

Use the add-model skill to start porting a new video model by staging reference code and initial weights, then proceed with conversion and verification workflows.

Frequently Asked Questions about add-model

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I port a new video model into FastVideo?▼

Porting a video model into FastVideo involves staging reference code, converting weights, and validating compatibility through structured phases including parity checks and configuration alignment.

What is weight conversion and parity testing for model integration?▼

Weight conversion and parity testing are validation phases in model integration that ensure a newly ported video model's outputs match the original reference code, confirming proper configuration and framework alignment.

How do I ensure configuration alignment when adding a new video generation model?▼

Configuration alignment is ensured through structured phases that handle weight conversion and configuration matching, validating that the new model integrates correctly into the FastVideo pipeline without manual guesswork.

Can I use this model porting workflow for novel video generation components?▼

Yes, this model porting workflow supports adding both new video model families and novel components into FastVideo, guiding you through staging reference code to final verification.

Why does my ported model fail parity checks during integration?▼

Parity checks fail during model integration when weight conversion or configuration alignment is incomplete, requiring you to validate compatibility and match settings against the reference code.

Related Skills

Explore More Software Engineering Skills→
FutureMLS-LabFutureMLS-Lab
556

sglang-diffusion-add-model

Integrate diffusion models and Diffusers pipelines into SGLang's multimodal runtime.

Official
Advanced
nidhalboumaiza-0nidhalboumaiza-0

Video Generation

Generate videos from text prompts or images using the z-ai-web-dev-sdk.

Community
Intermediate
lhllhl
38

check

Run three-stage Reality Check analyses on web sources with Python and RC-CLI tools.

Community
Advanced
invergent-aiinvergent-ai
807

new-model

Scaffold Surogate DSL blocks and HuggingFace weight mappings for new model architectures.

Official
Advanced