Hi AEON-7 team,
Thanks for publishing this project and the optimized DFlash container.
I am trying to run it on:
- Windows 11
- Docker Desktop with WSL2
- NVIDIA RTX 5090 Laptop GPU, 24GB VRAM
- 64GB RAM
- Host platform: linux/amd64
I have already downloaded the model volumes:
Qwen3.6-27B-AEON-Ultimate-Uncensored-Text-NVFP4-MTP-XS
Qwen3.6-27B-DFlash
When starting:
ghcr.io/aeon-7/vllm-aeon-ultimate-dflash:qwen36-v4
Docker reports:
The requested image's platform (linux/arm64) does not match the detected host platform (linux/amd64/v3)
I confirmed this with:
docker inspect ghcr.io/aeon-7/vllm-aeon-ultimate-dflash:qwen36-v4 --format '{{.Architecture}} {{.Os}}'
Output:
arm64 linux
Could you please provide either:
A linux/amd64 build of the DFlash-enabled container, or
Instructions for building it locally on linux/amd64?
I would like to run the NVFP4 + DFlash setup on an RTX 5090 Laptop GPU.
Thanks!
Hi AEON-7 team,
Thanks for publishing this project and the optimized DFlash container.
I am trying to run it on:
I have already downloaded the model volumes:
Qwen3.6-27B-AEON-Ultimate-Uncensored-Text-NVFP4-MTP-XSQwen3.6-27B-DFlashWhen starting: