Jarrel Seah

jarrelscy's activity

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

New activity in jarrelscy/GLM-5.3-Vision-NVFP4-BTX-hybrid 14 days ago

BTX hybrid v0 (untuned) - pipeline build

#2 opened 14 days ago by

jarrelscy

write probe

1

#1 opened 14 days ago by

jarrelscy

New activity in jarrelscy/GLM-5.2-NVFP4-AQLM-hybrid 2 months ago

I want 500k vision.

2

#1 opened 2 months ago by

taewan2002

New activity in jarrelscy/GLM-5.2-NVFP4-AQLM-hybrid-250k 2 months ago

Excellent Job! Question about the process.

4

#2 opened 2 months ago by

nevaback

Unknown quantization method: nvfp4_aqlm_hybrid.

1

#1 opened 3 months ago by

khalil2099

New activity in jarrelscy/Mixtral-8x22B-Instruct-v0.1-GPTQ-4bit over 2 years ago

I am running in vllm 0.4.1 with 4 x gpus 24gb (A10G 24gb) = 96gb and eager mode and I am still out of memory, how? it should fit (like 87gb vram)

1

#3 opened over 2 years ago by

orel12

KeyError: 'model.layers.45.block_sparse_moe.gate.g_idx'

5

#2 opened over 2 years ago by

tutu329

KeyError: 'model.layers.45.block_sparse_moe.gate.g_idx'

5

#2 opened over 2 years ago by

tutu329