Hugging Face's logo

Atomic-Germ
/
Qwopus3.5-9B-Coder-NPU2

Text Generation
English
qwen3_5
qwen3.5
coding
vision
multimodal
reasoning
tool-use
fastflowlm
q4nx
npu
conversational
Model card Files
xet
 Community
Qwopus3.5-9B-Coder-NPU2 / README.md
Atomic-Germ's picture
Atomic-Germ
Q4NX/FastFlowLM card: flm-add.py install + run/serve flow, kernels source, model info
b158caa verified
preview code
|
Raw
History Blame Contribute Delete
3.39 kB
metadata 
license: apache-2.0
language:
  - en
pipeline_tag: text-generation
tags:
  - qwen3.5
  - coding
  - vision
  - multimodal
  - reasoning
  - tool-use
  - fastflowlm
  - q4nx
  - npu
base_model:
  - Jackrong/Qwopus3.5-9B-v3.5

Qwopus3.5-9B-Coder - Q4NX for FastFlowLM (AMD Ryzen AI XDNA2)

Qwopus3.5-9B-v3.5, a vision-language coding model (reasoning, tool-use / function calling), converted to Q4NX for FastFlowLM. This variant ships vision Q4NX weights alongside the text weights.

What is Q4NX?

Q4NX is FastFlowLM's native packed-quantization format - a rearranged Q4_1 layout tuned for the NPU matrix engine's tile sizes and memory access patterns. It is not a GGUF file and it does not run on llama.cpp or Ollama; it is meant exclusively for the FastFlowLM engine on AMD Ryzen AI NPUs.

Requirements

  • FastFlowLM >= 0.9.45 (flm CLI)
  • AMD Ryzen AI processor with XDNA2 (NPU2) - Strix Point / Ryzen AI 300 series or later
  • Linux with the XRT NPU stack installed
  • ~16 GB of unified system memory (Q4NX weights + activations + KV cache)

Files

File Purpose
model.q4nx Quantized Q4NX weights
config.json FastFlowLM model configuration
tokenizer.json Tokenizer
tokenizer_config.json Special tokens and chat template
chat_template.jinja Chat template (optional)
vision_weight.q4nx Vision tower weights (multimodal input)
flm-add.py Installer script - registers this model with FastFlowLM

Install and run

This repository ships flm-add.py, a small installer that copies the model into the FastFlowLM user directory and registers the tag qwopus:9b. It never modifies the system FastFlowLM install.

# one-time environment (add these to ~/.bashrc)
export FLM_CONFIG_PATH="$HOME/.config/flm/model_list.json"
export FLM_XCLBIN_PATH="$HOME/.config/flm"

git lfs install
git clone https://huggingface.co/Atomic-Germ/Qwopus3.5-9B-Coder-NPU2
cd Qwopus3.5-9B-Coder-NPU2
python3 ./flm-add.py . --tag qwopus:9b --family qwen3.5
flm run qwopus:9b

Run python3 ./flm-add.py --help for all options. Without a clone, the same command works against the repo id directly:

python3 ./flm-add.py Atomic-Germ/Qwopus3.5-9B-Coder-NPU2 --tag qwopus:9b --family qwen3.5

Kernels

FastFlowLM's NPU kernels (xclbins) are closed source and are not shipped in this repository. flm-add.py links the kernels of the official qwen3.5:9b model (Qwen3.5-9B-NPU2), because this model shares the same engine family (qwen3.5) and architecture.

Serve (OpenAI-compatible) 

flm serve qwopus:9b --port 8080

curl http://127.0.0.1:8080/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -d '{"model":"qwopus:9b","messages":[{"role":"user","content":"Hello!"}],"max_tokens":256}'

Model

  • Registry tag: qwopus:9b
  • Engine family: qwen3.5
  • Kernel source: Qwen3.5-9B-NPU2
  • Context length: 262,144 tokens (from config)
  • model.q4nx size: 7.63 GB
  • Base model: Jackrong/Qwopus3.5-9B-v3.5
  • License: apache-2.0

Original model card

See the upstream model card for training details, benchmarks, and upstream usage. This repository only contains the Q4NX conversion for FastFlowLM.