yitong li

yitongl

3 5

AI & ML interests

None yet

Recent Activity

new activity 28 days ago

Efficient-Large-Model/StreamRender-H3:Document original upstream model download links

new activity 28 days ago

Efficient-Large-Model/StreamRender-H3:Complete runtime checkpoints and remove native video VAE dependency

new activity 28 days ago

Efficient-Large-Model/StreamRender-H3:Stream-Render: release trained renderer and causal decoder

View all activity

Organizations

yitongl 's Spaces 8

Kitchen Agent Round 1 Trace

🔎

93 trace records with faithful JSON and Chinese annotations

Fal vs Sol Blind Test

🎬

Rank video quality in a blind test

H3 FP4 Prefix-W16 Compare

🎬

Compare video outputs across model settings

Tae Test Prompts

👀

Explore and compare model video outputs across training steps

Latent Downsampler Online Evaluation

🎞

Step-0/200 latent downsampler video comparison

MiniMax-H3 FBCache Study

⚡

Seven-way MiniMax-H3 FBCache visual comparison.

LTX-2.5 ten prompts three ways

🎬

Compare three video generation configurations side by side

Video Opt Ablation

💻

yitong li

AI & ML interests

Recent Activity

Organizations

yitongl 's Spaces 8 Sort: Recently updated

Kitchen Agent Round 1 Trace

Fal vs Sol Blind Test

H3 FP4 Prefix-W16 Compare

Tae Test Prompts

Latent Downsampler Online Evaluation

MiniMax-H3 FBCache Study

LTX-2.5 ten prompts three ways

Video Opt Ablation

yitongl 's Spaces 8