---
license: other
language:
- en
pipeline_tag: text-generation
tags:
- fastflowlm
- q4nx
- npu
- qwen3.5
- 9b
- minicpm
base_model:
- ornith-ai/Ornith-1.0-9B
base_model_relation: quantized
quantized_by: Atomic-Germ
library_name: q4nx
---

# Ornith-1.0-9B - Q4NX for FastFlowLM (AMD Ryzen AI XDNA2)

Ornith-1.0-9B is converted to Q4NX for hardware-accelerated inference with FastFlowLM on AMD Ryzen AI NPUs.

## 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](https://fastflowlm.com) engine on AMD Ryzen AI NPUs.

## Requirements

- FastFlowLM >= 0.9.46 (`flm` CLI)
- AMD Ryzen AI processor with **XDNA2 (NPU2)** - Strix Point / Ryzen AI 300
  series or later
- Linux with the XRT NPU stack installed
- ~15 GB of unified system memory (Q4NX weights + activations + KV cache)

## Files

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

## Install and run

This repository works with `flm-add`, a small installer that copies the model
into the FastFlowLM user directory and registers the tag `minicpm4.6:0.8b`. It never
modifies the system FastFlowLM install.

`pip install flm-add` or `uv tool install flm-add`

```bash
uv tool install flm-add
flm-add Atomic-Germ/Ornith-1.0-9B-NPU2 --family qwen3.5
FLM_XCLBIN_PATH="$HOME/.config/flm FLM_CONFIG_PATH="$HOME/.config/flm/model_list.json flm run ornith:9b
```

## Kernels

FastFlowLM's NPU kernels (xclbins) are closed source and are not shipped in this repository. This model uses the **`qwen3.5`** engine family and is shape-identical to the official **`qwen3.5:9b`** model (`Qwen3.5-9B-NPU2`). Point the runtime's xclbin path at the matching `xclbins` directory (or ship your own) before running.

## Serve (OpenAI-compatible)

```bash
flm serve ornith:9b --port 8080
```

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

## Model

- Registry tag: `ornith:9b`
- Engine family: `qwen3.5`
- Kernel source: official `qwen3.5:9b` (`Qwen3.5-9B-NPU2`)
- Context length: 262,144 tokens (from config)
- Hidden size: 4096
- Layers: 32
- Intermediate size: 12288
- Vocabulary: 248320
- `model.q4nx` size: 7.11 GB
- Base model: [ornith-ai/Ornith-1.0-9B](https://huggingface.co/ornith-ai/Ornith-1.0-9B)
- License: other

## Original model card

See the upstream model card for training details, benchmarks, and upstream
usage. This repository only contains the Q4NX conversion for FastFlowLM.
- Upstream card: [ornith-ai/Ornith-1.0-9B](https://huggingface.co/ornith-ai/Ornith-1.0-9B)