1 2

Shahnawaz Ahmed

swaze

AI & ML interests

None yet

Recent Activity

upvoted a collection about 19 hours ago

Cosmos-Reason2

reacted to JonnaMat's post with 👀 3 days ago

🚀 FlashHead: Efficient Drop-In Replacement for the Classification Head in Language Model Inference 🔎 Check out our latest FlashHead-enabled model: https://huggingface.co/embedl/Cosmos-Reason2-2B-W4A16-Edge2-FlashHead 🧩 Seamless integration with vllm: ``` 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 \ embedl/vllm:latest-jetson-orin-flashhead \ vllm serve "embedl/Cosmos-Reason2-2B-W4A16-Edge2-FlashHead" \ --max-model-len 8192 \ --gpu-memory-utilization 0.75 \ --max-num-seqs 2 \ --trust-remote-code ```

liked a Space 6 days ago

jane-street/droppedaneuralnet

View all activity

Organizations

upvoted a collection about 19 hours ago

Cosmos-Reason2

Collection

nvidia/Cosmos-Reason2 multi-modal reasoning models optimized by Embedl. • 8 items • Updated 1 day ago • 4

reacted to JonnaMat's post with 👀 3 days ago

Post

5864

🚀 FlashHead: Efficient Drop-In Replacement for the Classification Head in Language Model Inference

🔎 Check out our latest FlashHead-enabled model: embedl/Cosmos-Reason2-2B-W4A16-Edge2-FlashHead

🧩 Seamless integration with vllm:

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 \
  embedl/vllm:latest-jetson-orin-flashhead \
  vllm serve "embedl/Cosmos-Reason2-2B-W4A16-Edge2-FlashHead" \
    --max-model-len 8192 \
    --gpu-memory-utilization 0.75 \
    --max-num-seqs 2 \
    --trust-remote-code