# Can the Apple M3 Ultra (512 GB) run Qwen3.8-Flash-Next?

> Yes. The Apple M3 Ultra (512 GB) runs Qwen3.8-Flash-Next fully in unified memory at FP16 / BF16 (about 347 GB of 384 GB), at roughly 3.7 tokens/s.

Apple M3 Ultra (512 GB): 512 GB (384 GB usable), 819 GB/s. Estimates at an 8K context.

## Every quantisation

| Quant | Quality | ≈ GB | Fit | Est. tok/s |
|---|---|---|---|---|
| Q2_K | Heavy loss | 59.12 | fits | 22 |
| Q3_K_M | Noticeable loss | 87.28 | fits | 15 |
| Q4_K_M | Small loss | 107.96 | fits | 12 |
| Q5_K_M | Near-lossless | 126.44 | fits | 10 |
| Q6_K | Near-lossless | 146.25 | fits | 8.9 |
| Q8_0 | Lossless | 185.32 | fits | 6.9 |
| F16 (pick) | Lossless | 347.46 | fits | 3.7 |

## Run it

From the Ollama library (ollama):

```bash
ollama run qwen3.8-flash-next
```

BF16 straight from Hugging Face (ollama):

```bash
ollama run hf.co/unsloth/Qwen3.8-Flash-Next-GGUF:BF16
```

OpenAI-compatible server with a web UI (llama.cpp):

```bash
llama-server -hf unsloth/Qwen3.8-Flash-Next-GGUF:BF16
```

Download with the lms CLI (lmstudio):

```bash
lms get unsloth/Qwen3.8-Flash-Next-GGUF@BF16
```

Ask about an image (transformers):

```python
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="Qwen/Qwen3.8-Flash-Next", dtype="auto", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
            {"type": "text", "text": "Describe this image."},
        ],
    }
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])
```


## Vision models that fit comfortably

- [MiniMaxAI/MiniMax-M3.1-Flash](https://selfhost.directory/models/minimaxai/minimax-m3.1-flash/on/apple-m3-ultra-512gb): Q8_0 ≈3.49 GB, est. 1.9 tok/s
- [zai-org/GLM-5.3-Flash](https://selfhost.directory/models/zai-org/glm-5.3-flash/on/apple-m3-ultra-512gb): Q8_0 ≈356.44 GB, est. 23 tok/s
- [Qwen/Qwen3.5-122B-A10B](https://selfhost.directory/models/qwen/qwen3.5-122b-a10b/on/apple-m3-ultra-512gb): Q8_0 ≈128.25 GB, est. 42 tok/s
- [Qwen/Qwen3-VL-235B-A22B-Instruct](https://selfhost.directory/models/qwen/qwen3-vl-235b-a22b-instruct/on/apple-m3-ultra-512gb): Q8_0 ≈246.38 GB, est. 19 tok/s
- [nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16](https://selfhost.directory/models/nvidia/nvidia-nemotron-3.5-super-vl-120b-a12b-bf16/on/apple-m3-ultra-512gb): Q8_0 ≈130.45 GB, est. 35 tok/s
- [tencent/Hy4-flash](https://selfhost.directory/models/tencent/hy4-flash/on/apple-m3-ultra-512gb): Q8_0 ≈231.97 GB, est. 1.9 tok/s

## Other GPUs that run it

- [Apple M4 Max 40-core (128 GB)](https://selfhost.directory/models/qwen/qwen3.8-flash-next/on/apple-m4-max-128gb-40c): Q3_K_M, est. 10 tok/s
- [RTX PRO 6000 Blackwell](https://selfhost.directory/models/qwen/qwen3.8-flash-next/on/rtx-pro-6000-blackwell): Q3_K_M, est. 33 tok/s
- [Ryzen AI Max+ 395 (128 GB)](https://selfhost.directory/models/qwen/qwen3.8-flash-next/on/ryzen-ai-max-395-128gb): Q3_K_M, est. 4.7 tok/s
- [Apple M2 Ultra (192 GB)](https://selfhost.directory/models/qwen/qwen3.8-flash-next/on/apple-m2-ultra-192gb): Q5_K_M, est. 10 tok/s

---
Model: https://selfhost.directory/models/qwen/qwen3.8-flash-next. GPU: https://selfhost.directory/models/gpus/apple-m3-ultra-512gb. Source: https://selfhost.directory/models/qwen/qwen3.8-flash-next/on/apple-m3-ultra-512gb