Kaizhao Liang's picture

Kaizhao Liang PRO

kz919

·

https://kyleliang919.github.io/

AI & ML interests

Search = AGI?

Recent Activity

liked a dataset 4 days ago

laion/laion400m

liked a Space 10 days ago

lllyasviel/IC-Light

liked a dataset 12 days ago

timm/imagenet-1k-wds

View all activity

Organizations

Posts 7

Post

1549

Mini-QwQ an edge device friendly reasoning model distilled from QwQ-32B
🤗: kz919/QwQ-0.5B-Distilled-SFT
🇬 🇬 🇺 🇫: kz919/QwQ-0.5B-Distilled-SFT-gguf
🤖: kz919/Mini-QwQ

Post

1511

Just for the meme.

But the clear lesson I learnt from building these demos are, the more powerful the underlying base model is, the closer you will get to GPT4o1. CoT is nothing more than simply inducing the latent reasoning capability from the model.

kz919/GPT4-O1-Proximas

Papers 10

arxiv:2411.16085

arxiv:2405.07518

arxiv:2404.00438

arxiv:2310.05898

spaces 4

Mini QwQ

Qwen 0.5B but QwQ

GPT4 O1 Proximas

Generate detailed step-by-step answers

Can You Beat 405B In Chess

Play chess against an AI

Persona AI

Chat with customizable AI characters

models 4

kz919/QwQ-0.5B-Distilled

Text Generation • Updated Jan 11 • 62 • 5

kz919/QwQ-0.5B-Distilled-SFT

Text Generation • Updated Jan 11 • 306 • 23

kz919/QwQ-0.5B-Distilled-SFT-gguf

Text Generation • Updated Jan 10 • 576 • 7

kz919/QwQ-0.5B-Distilled-gguf

Updated Jan 3 • 87 • 2

datasets 4

kz919/pg19-llama3-tokenized

Viewer • Updated Jul 20, 2024 • 28.6k • 159

kz919/pg19-test-llama3-tokenized

Viewer • Updated Jul 19, 2024 • 100 • 66

kz919/coe_router_data

Viewer • Updated Feb 29, 2024 • 1.4k • 104 • 2

kz919/mmlu-auxiliary-train-auto-labelled

Viewer • Updated Jan 31, 2024 • 99.8k • 290 • 4