Sign inSign up

avarok/atlas-alpha2

By avarok

•Updated 7 months ago

Image
0

4.5K

avarok/atlas-alpha2 repository overview

⁠Atlas Spark — Alpha 2

Image: docker pull avarok/atlas-alpha2 Hardware: NVIDIA DGX Spark GB10

⁠Models

ModelHuggingFace IDTypeMTPtok/s
35B MoEKbenkhaled/Qwen3.5-35B-A3B-NVFP4SSM+Attn+MoEYes~130
80B MoEnvidia/Qwen3-Next-80B-A3B-Instruct-NVFP4SSM+Attn+MoEYes~100
VL-30Big1/Qwen3-VL-30B-A3B-Instruct-NVFP4Attn+MoE (Vision)No~100
Nemotron-H 30Bnvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4Mamba-2+MoE+AttnNo~100
27B DenseKbenkhaled/Qwen3.5-27B-NVFP4SSM+Attn (Dense)No~14
122B MoESehyo/Qwen3.5-122B-A10B-NVFP4SSM+Attn+MoEYes~50

Download models: hf download <HuggingFace ID>

⁠Run Commands

35B (recommended):

sudo docker run -d --name atlas --gpus all --ipc=host --network host \
  -v ~/.cache/huggingface:/root/.cache/huggingface \
  avarok/atlas-alpha2 serve Kbenkhaled/Qwen3.5-35B-A3B-NVFP4 \
    --port 8888 --kv-cache-dtype nvfp4 --gpu-memory-utilization 0.88 \
    --scheduling-policy slai --max-seq-len 8192 --max-batch-size 16 \
    --speculative --mtp-quantization nvfp4

⁠License

This project is licensed under the GNU Affero General Public License (AGPL)⁠. If you modify or deploy this image as part of a network-accessible service, you must make the complete corresponding source code available to its users under the same license. This applies whether you interact with the software directly or through a container. We chose AGPL specifically to ensure source availability in an era where binary-only distribution no longer provides meaningful obscurity.

For the full license text, see the LICENSE file included in this repository.

Tag summary

Content type

Image

Digest

sha256:880d25e54…

Size

1.8 GB

Last updated

7 months ago

docker pull avarok/atlas-alpha2