Hugging Face's logo
Activity Feed

AI & ML interests

Canada Quant Labs — Canada's open-weight model lab. We train, quantize, and ship sovereign reference models for regulated industries (legal, medical, defence, finance) on a DGX B300 at Equinix Vancouver. Upstream contributors to vLLM and llm-compressor. Recipes: W4A16, NVFP4, MXFP4. Built in Victoria, BC. partnerships@cql.ca · cql.ca

Recent Activity
pastapaul  updated a model about 1 month ago
canada-quant/hy3-w4a16-mtp
pastapaul  updated a dataset about 1 month ago
canada-quant/hy3-w4a16-mtp-calibration
pastapaul  published a dataset about 1 month ago
canada-quant/hy3-w4a16-mtp-calibration
View all activity
pastapaul 
updated a model about 1 month ago
pastapaul 
updated a dataset about 1 month ago
pastapaul 
published a dataset about 1 month ago
pastapaul 
updated a Space about 2 months ago
pastapaul 
published a model about 2 months ago
pastapaul 
in canada-quant/GLM-5.2-W4A16-MTP about 2 months ago

Loading the model

4
#1 opened 2 months ago by
madeby561
pastapaul 
updated a model about 2 months ago
pastapaul 
in canada-quant/GLM-5.2-W4A16-MTP 2 months ago

Can you run NVFP4 + Marlin?

5
#2 opened 2 months ago by
einsteiner1983
pastapaul 
published a model 2 months ago
pastapaul 
in canada-quant/DeepSeek-V4-Flash-NVFP4-FP8-MTP 3 months ago

Docker Image

8
#1 opened 3 months ago by
mtcl
pastapaul 
in canada-quant/DeepSeek-V4-Flash-W4A16-FP8 3 months ago

Compatible version with Ampere? SM8.6

1
#6 opened 3 months ago by
bullerwins

Small patch to fix MTP on current vLLM (8th of June, 26)

7
#8 opened 3 months ago by
dnhkng
pastapaul 
in canada-quant/DeepSeek-V4-Flash-W4A16-FP8-MTP 3 months ago

MTP BF16 draft weights may need explicit quant_config exclusion for vLLM

2
#2 opened 3 months ago by
yangsiqt2

vLLM compressed-tensors loader fails on shared_experts.gate_up_proj — missing quant target for fused shared-expert projection?

3
#1 opened 3 months ago by
ajdoosh
pastapaul 
updated a model 3 months ago
pastapaul 
updated a dataset 3 months ago
pastapaul 
published a dataset 3 months ago
pastapaul 
updated a model 3 months ago
pastapaul 
in canada-quant/DeepSeek-V4-Flash-W4A16-FP8 3 months ago

2026-05-26 dual-spark bench: GSM8K 94-98%, AIME 4/5, concurrency wall 12→125 tok/s
#7 opened 3 months ago by
pastapaul
pastapaul 
in canada-quant/DeepSeek-V4-Pro-NVFP4-FP8-MTP 3 months ago

Lossless?

👀 1
1
#1 opened 3 months ago by
lukealonso