README.md

---
license: apache-2.0
base_model:
- Qwen/Qwen3.6-35B-A3B-FP8
pipeline_tag: image-text-to-text
tags:
- radiance
- rocm
- amd
- rdna4
- moe
- fp8
- speculative-decoding
- mtp
- vision
---

# Qwen3.6-35B-A3B FP8 — radiance container

[Qwen/Qwen3.6-35B-A3B-FP8](https://huggingface.co/Qwen/Qwen3.6-35B-A3B-FP8) as a single `.rad`
container for the **radiance** inference engine (AMD RDNA4, ROCm), with its MTP head for
speculative decoding and its vision tower.

| | |
|---|---|
| File | `qwen3.6-35b-a3b-fp8.rad` — 34.63 GiB |
| Weights | the published FP8 checkpoint as it ships: every linear, the 256 routed experts included, E4M3 with a scale per 128×128 block; the lm_head made block FP8 by the recipe |
| Speculator | the model's MTP head, which drafts with a 2-bit copy of the lm_head |
| Vision | the 27-block vision tower, bf16: images and video in chat requests |
| Context | 262,144 tokens trained; 200K tested |

## Serve

The whole model fits two 32 GB cards.

```sh
radiance --model qwen3.6-35b-a3b-fp8.rad --tp 2 --max-model-len 200000 --kv-cache-dtype fp8 \
    --max-num-seqs 8 --host 0.0.0.0 --port 8000
```

The server speaks the OpenAI API (`/v1/chat/completions`, `/v1/completions`), with tool calls and
structured output, and `image_url` / `video_url` parts in chat
messages. The MTP depth is chosen automatically (`--num-speculative-tokens N` states one, `0` turns
speculation off).

Built and tested on 2× Radeon AI PRO R9700 (gfx1201).

## How this file was made

```sh
rad-convert Qwen/Qwen3.6-35B-A3B-FP8 --recipe q36-35b-a3b-fp8.recipe -o qwen3.6-35b-a3b-fp8.rad
```

The recipe (`q36-35b-a3b-fp8.recipe` in this repository); everything it does not name is the
checkpoint's own:

```
output.weight rtn codes=fp8_e4m3 block=128x128 scale=bf16
mtp.draft_head.weight rtn codes=u2 zero=u8 group=128 scale=f16
```

## License

Apache 2.0, as the base model; see `LICENSE`.