Hannes von Essen's picture

Hannes von Essen

HProg

embedl

·

AI & ML interests

Deep Learning Researcher at Embedl

Recent Activity

reacted to JonnaMat's post with 🔥 about 12 hours ago

🤯 Edge-Grade Vision Reasoning. Now Practically Lossless. 🤯 Introducing 👉 https://huggingface.co/embedl/Cosmos-Reason2-2B-W4A16-Edge2 Optimized for Jetson Orin Nano Super and AGX Orin https://huggingface.co/nvidia . 🚄 Try it out on Jetson (image+video+text): ``` docker run --rm -it \ --network host \ --shm-size=8g \ --ulimit memlock=-1 \ --ulimit stack=67108864 \ --runtime=nvidia \ --name=vllm-serve \ -e HF_TOKEN=hf_*** \ -e HF_HOME=/root/.cache/huggingface \ ghcr.io/nvidia-ai-iot/vllm:latest-jetson-orin \ vllm serve "embedl/Cosmos-Reason2-2B-W4A16-Edge2" \ --max-model-len 8192 \ --gpu-memory-utilization 0.75 \ --max-num-seqs 2 ``` 🤓 What is Edge2? Most weights → INT4 | Activations → FP16 | Select sensitive layers → kept in FP16. Edge2 preserves precision where it matters most; while keeping the model small and fast enough for edge GPUs. 😎

posted an update about 23 hours ago

Cosmos-Reason2-2B on Nano Super Hi! Today, me and my team is releasing a version of Cosmos-Reason2-2B that is quantized so that it fits on the NVIDIA Jetson Orin Nano Super. We managed to find a mixed precision configuration such that it maintains virtually the same accuracy as the unquantized model while being able to run really efficiently on the Nano Super and other edge devices :) https://huggingface.co/embedl/Cosmos-Reason2-2B-W4A16-Edge2

reacted to JonnaMat's post with 🔥 1 day ago

⚡ Blackwell-native Vision Reasoning at the edge ⚡ Released a NVFP4A16-variant of nvidia/Cosmos-Reason2-2B: https://huggingface.co/embedl/Cosmos-Reason2-2B-NVFP4A16 💖 Optimized for Blackwell with minimal accuracy drop compared to its FP16 counterpart. Thorough on-device benchmarks on AGX Thor in the modelcard. 🤓 📊 Try it out: ``` docker run --rm -it \ --network host \ --shm-size=8g \ --ulimit memlock=-1 \ --ulimit stack=67108864 \ --runtime=nvidia \ --name=vllm-serve \ -e HF_TOKEN=hf_*** \ -e HF_HOME=/root/.cache/huggingface \ nvcr.io/nvidia/vllm:26.01-py3 \ vllm serve "embedl/Cosmos-Reason2-2B-NVFP4A16" \ --host 0.0.0.0 \ --port 8000 \ --tensor-parallel-size 1 \ --max-model-len 16384 \ --gpu-memory-utilization 0.9 ```

View all activity

Organizations

HProg 's datasets

None public yet