You can deploy this model and start earning money today!

blockblockblock/
VicUnlocked-alpaca-30b-exl3-3.0bpw

Name: VicUnlocked-alpaca-30b-exl3-3.0bpw
Availability: InStock

Aeala/VicUnlocked-alpaca-30b

quantized

other

3-bit

Views

No views yet

VicUnlocked-alpaca-30b

_{EXL3 · 3.0 bpw · 12.7 GB · Llama · 60 layers}

base model

quantized by

[!NOTE] An ExLlamaV3 build of Aeala/VicUnlocked-alpaca-30b at 3.0 bits per weight. See Quants for sibling repos at other bit‑widths.

Quants

BPW	Head bits	Calibration rows	Size	Status
2.5	8	128	10.7 GB	link
3.0	8	128	12.7 GB	`this repo`

Inference

Loader	Use it for
TabbyAPI	OpenAI‑compatible HTTP server. Drop‑in for OpenAI clients.
text‑generation‑webui	Local chat UI. Pick the ExLlamaV3 loader from the model dropdown.
ExLlamaV3	Direct Python API for embedding the model in your own code or pipeline.

[!TIP] VRAM at 3.0 bpw: ~12.7 GB weights + ~2 GB context. Comfortable on a single 16 GB card; tight on 12 GB with reduced context.

Download

bash

pip install -U huggingface_hub

hf download \
  blockblockblock/VicUnlocked-alpaca-30b-exl3-3.0bpw \
  --local-dir ./VicUnlocked-alpaca-30b-exl3-3.0bpw

Quantization recipe _{(advanced — embedded in quantization_config.json)}

Setting	Value
Format	`EXL3`
Bits per weight	`3.0`
Head bits	`8`
Calibration rows	`128`
Codebook	`MCG`
Out‑scales	`always`

Loaded automatically by every ExLlamaV3 loader; reproduced here for searchability.

License & use

[!IMPORTANT] Use and license follow the base model. Quantization adds no additional restrictions. Refer to the upstream repository for terms, citation, and safety documentation.

_{Quantized with BlockQuant · convention {org}/{model}-exl3-{bpw}bpw}