CoolFace
Modelpublic

drawais/Granite-3.3-8B-Instruct-AWQ-INT4

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face
0likes158downloads
Model Card

Granite-3.3-8B-Instruct-AWQ-INT4

INT4 weight-only quantization of `ibm-granite/granite-3.3-8b-instruct`.

IBM Granite 3.3 8B-Instruct in INT4. About 5 GB on disk. Runs on an 8 GB consumer GPU.

PropertyValue
Base modelibm-granite/granite-3.3-8b-instruct
QuantizationINT4 weight-only
Approx. on-disk size~4.5 GB
LicenseApache License, Version 2.0
LanguagesEnglish

Load (vLLM)

bash
vllm serve drawais/Granite-3.3-8B-Instruct-AWQ-INT4 \
  --max-model-len 32768 \
  --gpu-memory-utilization 0.94
python
from vllm import LLM, SamplingParams
llm = LLM(model="drawais/Granite-3.3-8B-Instruct-AWQ-INT4", max_model_len=32768)
print(llm.generate(["Hello!"], SamplingParams(max_tokens=128))[0].outputs[0].text)

Footprint

~4.5 GB on disk. Recommended VRAM: enough headroom for KV cache.

License & attribution

This artifact is a derivative work of `ibm-granite/granite-3.3-8b-instruct`, released by its original authors under the Apache License, Version 2.0.

This artifact is distributed under the same license. The full license text is included in `LICENSE`, and required attribution is in `NOTICE`.

License text: https://www.apache.org/licenses/LICENSE-2.0 Source model: https://huggingface.co/ibm-granite/granite-3.3-8b-instruct