Yi Cui PRO

onekq's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes

updated a Space about 1 month ago

Running

3

🥇

WebApp1K Models Leaderboard

posted an update about 2 months ago

view post

Post

565

October version of Claude 3.5 lifts SOTA (set by its June version) by 7 points.
onekq-ai/WebApp1K-models-leaderboard

Closed sourced models are widening the gap again.

Note: Our frontier leaderboard now uses double test scenarios because the single-scenario test suit has been saturated.

👍

1

1

+

Reply

New activity in onekq-ai/WebApp1K-models-leaderboard about 2 months ago

All the clickable links are not accessible...

2

#3 opened about 2 months ago by

zhiminy

The leaderboard is not displaying well...

3

#2 opened about 2 months ago by

zhiminy

Quick fix.

#4 opened about 2 months ago by

John6666

updated a model 2 months ago

onekq-ai/starcoder2-3b-instruct-v0.1

Text Generation • Updated Oct 19 • 22

updated 2 collections 2 months ago

QLora-ready Coding Models

Collection

For Finetuning. GPU is needed for both quantization and inference. • 9 items • Updated Oct 19

Ollama-ready Coding Models

Collection

For inference. CPU is enough for both quantization and inference. • 14 items • Updated Oct 19 • 2

updated a model 2 months ago

onekq-ai/DeepSeek-Coder-V2-Lite-Base-bnb-4bit

Text Generation • Updated Oct 19 • 15

posted an update 2 months ago

view post

Post

1846

I'm now working on finetuning of coding models. If you are GPU-hungry like me, you will find quantized models very helpful. But quantization for finetuning and inference are different and incompatible. So I made two collections here.

Inference (GGUF, via Ollama, CPU is enough)
onekq-ai/ollama-ready-coding-models-67118c3cfa1af2cf04a926d6

Finetuning (Bitsandbytes, QLora, GPU is needed)
onekq-ai/qlora-ready-coding-models-67118771ce001b8f4cf946b2

For quantization, the inference models are far more popular on HF than finetuning models. I use https://huggingface.co/QuantFactory to generate inference models (GGUF), and there are a few other choices.

But there hasn't been such a service for finetuning models. DIY isn't too hard though. I made a few myself and you can find the script in the model cards. If the original model is small enough, you can even do it on a free T4 (available via Google Colab).

If you know a (small) coding model worthy of quantization, please let me know and I'd love to add it to the collections.

🔥

6

6

+

Reply

updated a collection 2 months ago

Ollama-ready Coding Models

Collection

For inference. CPU is enough for both quantization and inference. • 14 items • Updated Oct 19 • 2

updated 3 models 2 months ago

onekq-ai/starcoder2-3b-bnb-4bit

Text Generation • Updated Oct 18 • 56

onekq-ai/starcoder2-7b-bnb-4bit

Text Generation • Updated Oct 18 • 10

onekq-ai/starcoder2-15b-bnb-4bit

Text Generation • Updated Oct 18 • 13 • 1

updated a collection 2 months ago

Ollama-ready Coding Models

Collection

For inference. CPU is enough for both quantization and inference. • 14 items • Updated Oct 19 • 2