Skip to content
Log in

Qwen/Qwen3.8-27B

huggingface_int4_embedl-flashhead-mtp_tensorrt-edgellm
Qwen/Qwen3.8-27B architecture graph
Qwen/Qwen3.8-27B architecture graph
  • huggingface
  • AGX Thor
Labels
  • Image
  • Text
  • VLM
  • Generation
Build
Embedl-optimized
Precision
W4A16
Runtime
TensorRT Edge-LLM 0.10.1 with Embedl optimizations
Context length
4,096 tokens

Alibaba's 27B Qwen3.8 model for coding, research and long multi-step agent tasks, with images and video.

Files

Image Size

Model

docker.io/embedl/model-artifacts:qwen3-8-27b.huggingface_int4_embedl-flashhead-mtp_tensorrt-edgellm.039c3dd747d129b6

 20.29 GiB

Runtime

docker.io/embedl/models:runtimes.tensorrt-edgellm.arm64.2026101815014ee0

 1.01 GiB