Downloads · 30 days
27
3% of all-time downloads
blockblockblock/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-exl3-3.0bpw
Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-exl3-3.0bpw is a text generation model from blockblockblock. Use it when you need the model to write or continue text. It is set up for exllamav3. The card lists the license as other.
Downloads · 30 days
27
3% of all-time downloads
All-time downloads
856
Public
Parameters
7.6B
15.3 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors15.3 GB · 100%
How the weights are stored.
I166.6B · 86%
From the Hugging Face model README
<sub><code>EXL3</code> · <b>3.0 bpw</b> · 15.3 GB · Mixture‑of‑Experts · 48 layers × 256 experts</sub>
<br/> </div>[!NOTE] An ExLlamaV3 build of
lordx64/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilledat 3.0 bits per weight — the tightest fit — sized for 16 GB consumer cards while leaving usable context room. See Quants for sibling repos at other bit‑widths or browse the collection.
| BPW | Head bits | Calibration rows | Size | Status |
|---|---|---|---|---|
| 3.0 | 8 | 128 | 15.3 GB | <kbd>this repo</kbd> |
| 4.0 | 8 | 128 | 19.5 GB | link |
| 4.5 | 8 | 250 | 21.6 GB | link |
| 5.0 | 8 | 250 | 23.7 GB | link |
| 6.0 | 8 | 250 | 27.9 GB | link |
[!TIP] VRAM at 3.0 bpw: weights on disk + ~2 GB context overhead. Fits a 16 GB card with workable context, comfortable on 24 GB with very long context.
pip install -U huggingface_hub
hf download \
blockblockblock/Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-exl3-3.0bpw \
--local-dir ./Qwen3.6-35B-A3B-Claude-4.7-Opus-Reasoning-Distilled-exl3-3.0bpw
<details>
<summary><b>Quantization recipe</b> <sub>(advanced — embedded in <code>quantization_config.json</code>)</sub></summary>
<br/>
| Setting | Value |
|---|---|
| Format | EXL3 |
| Bits per weight | 3.0 |
| Head bits | 8 |
| Calibration rows | 128 |
| Codebook | MCG |
| Out‑scales | always |
| Parallel mode | enabled (MoE expert batching) |
Loaded automatically by every ExLlamaV3 loader; reproduced here for searchability.
</details>[!IMPORTANT] Use and license follow the base model. Quantization adds no additional restrictions. Refer to the upstream repository for terms, citation, and safety documentation.