Hugging Face's logo
Activity Feed

AI & ML interests

OpenSource and AI

Recent Activity
dsikka  new activity 2 days ago
RedHatAI/Qwen3.6-35B-A3B-NVFP4:Great quant!!
MeganEFlynn  updated a model 3 days ago
RedHatAI/gemma-4-31B-it-speculator.dflash
MeganEFlynn  new activity 4 days ago
RedHatAI/gemma-4-31B-it-speculator.dflash:Amazing
View all activity

Papers

S2D2: Fast Decoding for Diffusion LLMs via Training-Free Self-Speculation

View all Papers
dsikka 
in RedHatAI/Qwen3.6-35B-A3B-NVFP4 2 days ago

Great quant!!

9
#6 opened 4 days ago by
tasticleeze
MeganEFlynn 
updated a model 3 days ago
MeganEFlynn 
in RedHatAI/gemma-4-31B-it-speculator.dflash 4 days ago

Amazing

2
#2 opened 5 days ago by
ehartford
dsikka 
in RedHatAI/Qwen3.6-35B-A3B-NVFP4 4 days ago

Regarding the correctness of the int4 quantization script

1
#5 opened 5 days ago by
traphix
dsikka 
in RedHatAI/Qwen3.6-35B-A3B-NVFP4 5 days ago

config.json ignore-list needs linear_attn patterns — model produces garbage otherwise

❤️ 1
3
#4 opened 7 days ago by
cghart123
fynnsu 
in RedHatAI/gemma-4-26B-A4B-it-speculator.eagle3 5 days ago

Not working in VLLM

3
#1 opened 8 days ago by
pachePizza
dsikka 
in RedHatAI/Qwen3.6-35B-A3B-NVFP4 5 days ago

Creation details?

1
#3 opened 7 days ago by
traphix
dsikka 
updated a model 5 days ago
dsikka 
in RedHatAI/Qwen3.6-35B-A3B-NVFP4 5 days ago

Sglang?

1
#2 opened 8 days ago by
jpsequeira

vllm-openai:cu130-nightly Error

➕ 4
3
#1 opened 9 days ago by
andynoodles
fynnsu 
in RedHatAI/gemma-4-26B-A4B-it-speculator.eagle3 5 days ago

hoping to add multi-language support

👀 1
2
#2 opened 7 days ago by
TIN-HF
dsikka 
updated a model 8 days ago
dsikka 
updated a collection 8 days ago
MeganEFlynn 
published a model 8 days ago
fynnsu 
in RedHatAI/Qwen3.5-397B-A17B-FP8-dynamic 8 days ago

Creation details?

1
#8 opened about 1 month ago by
traphix
fynnsu 
in RedHatAI/Qwen3-8B-Thinking-speculator.eagle3 8 days ago

Update vllm serve command with full HF name for Qwen/Qwen3-8B
#1 opened 8 days ago by
reed-meyerson
bdellabe 
in RedHatAI/gemma-4-31B-it-NVFP4 8 days ago

Error KeyError: 'layers.0.experts.0.down_proj.input_global_scale' when running on vllm

1
#5 opened 8 days ago by
pachePizza
reed-meyerson 
in RedHatAI/Qwen3-8B-Thinking-speculator.eagle3 8 days ago

Update vllm serve command with full HF name for Qwen/Qwen3-8B
#1 opened 8 days ago by
reed-meyerson
dsikka 
published a model 9 days ago
fynnsu 
updated a collection 9 days ago