embedl/Cosmos-Reason2-2B-W4A16-Edge2-FlashHead
Image-Text-to-Text • 2B • Updated • 921 • 9
Jonna Matthiesen
JonnaMat
AI & ML interests
None yet
Recent Activity
repliedto their post 1 day ago
🚗 The reasoning backbone quadruples from 8B to 32B , while the action expert remains at 2.3B!
👀 We took a closer look at the architectural evolution from https://huggingface.co/nvidia/Alpamayo-1.5-10B to https://huggingface.co/nvidia/Alpamayo2-Super .
Read the analysis here:
https://huggingface.co/blog/JonnaMat/alpamayo2-super
Our analysis explores some implications of this design choice, especially from a distillation perspective where keeping the expert compact could be key for efficient deployment. 🧠 posted an update 2 days ago
🚗 The reasoning backbone quadruples from 8B to 32B , while the action expert remains at 2.3B!
👀 We took a closer look at the architectural evolution from https://huggingface.co/nvidia/Alpamayo-1.5-10B to https://huggingface.co/nvidia/Alpamayo2-Super .
Read the analysis here:
https://huggingface.co/blog/JonnaMat/alpamayo2-super
Our analysis explores some implications of this design choice, especially from a distillation perspective where keeping the expert compact could be key for efficient deployment. 🧠 published an article 2 days ago
Alpamayo 2 Super: The expert that didn't growOrganizations
liked a
model 4 months ago
liked a
model 5 months ago
liked a
Space 5 months ago
liked a
model 5 months ago
liked 8
models 8 months ago
embedl/Llama-3.2-3B-Instruct-FlashHead-W4A16
4B • Updated • 26 • 4
embedl/gemma-3-1b-it-FlashHead-W4A16
1B • Updated • 23 • 3
embedl/Llama-3.2-3B-Instruct-FlashHead
3B • Updated • 25 • 4
embedl/gemma-3-1b-it-FlashHead
1.0B • Updated • 23 • 3
embedl/Llama-3.2-1B-Instruct-FlashHead-W4A16
2B • Updated • 63 • 6
embedl/Qwen3-1.7B-FlashHead-W4A16
2B • Updated • 112 • 3
embedl/Qwen3-1.7B-FlashHead
2B • Updated • 146 • 3
embedl/Llama-3.2-1B-Instruct-FlashHead
1B • Updated • 164 • 4