tp's picture

12

tp

Onix22

·

AI & ML interests

None yet

Recent Activity

new activity about 16 hours ago

zetasepic/Rombo-LLM-V3.1-QWQ-32b-abliterate-6.5bpw-h8-exl2:you should make 4 or 4.25 bit quantization to make it usable on 24GB cards

new activity 18 days ago

nbeerbower/Dumpling-Qwen2.5-VL-7B:What it is made for?

new activity 27 days ago

huihui-ai/Qwen2-VL-7B-Instruct-abliterated:Request gguf

View all activity

Organizations

None yet

Onix22's activity

New activity in zetasepic/Rombo-LLM-V3.1-QWQ-32b-abliterate-6.5bpw-h8-exl2 about 16 hours ago

you should make 4 or 4.25 bit quantization to make it usable on 24GB cards

#1 opened about 16 hours ago by

New activity in nbeerbower/Dumpling-Qwen2.5-VL-7B 18 days ago

What it is made for?

#1 opened 19 days ago by

New activity in huihui-ai/Qwen2-VL-7B-Instruct-abliterated 27 days ago

Request gguf

#1 opened 5 months ago by

New activity in Apel-sin/deepseek-r1-distill-qwen-32B-abliterated-exl2 about 1 month ago

So where is the model I would like to try it?

#1 opened about 1 month ago by

New activity in TheBloke/CodeLlama-34B-GPTQ about 1 year ago

This model breaks laws of physic.

#4 opened about 1 year ago by

New activity in TheBloke/WizardLM-Uncensored-SuperCOT-StoryTelling-30B-SuperHOT-8K-GPTQ over 1 year ago

Does it work for anyone?

#1 opened over 1 year ago by

New activity in anon8231489123/gpt4-x-alpaca-13b-native-4bit-128g almost 2 years ago

Running out of memory with 12GB of VRAM on 3080TI

#39 opened almost 2 years ago by

If 4-bit = 8GB model would 8-bit = 16GB? And how much better would it be?

#40 opened almost 2 years ago by

New activity in tsumeone/llama-30b-supercot-3bit-128g-cuda almost 2 years ago

30B 3bit seems pretty sweet by the official evaluation

#1 opened almost 2 years ago by

New activity in GamerUntouch/LLaMa-Storytelling-4Bit almost 2 years ago

update is needed

#5 opened almost 2 years ago by

New activity in tsumeone/llama-30b-supercot-4bit-cuda almost 2 years ago

3bit version

#1 opened almost 2 years ago by