metadata
license: apache-2.0
base_model: Qwen/Qwen2.5-Coder-14B-Instruct
tags:
- quantized
- 4-bit
- int4
- qwen2.5
- coder
- code
language:
- en
pipeline_tag: text-generation
Qwen2.5-Coder-14B-Instruct-HQQ-INT4
INT4 quantization of Qwen/Qwen2.5-Coder-14B-Instruct. Calibration-free companion to drawais/Qwen2.5-Coder-14B-Instruct-AWQ-INT4.
Footprint
| Source params | 14B (code-specialized) |
| Quantized weights | ~9.5 GB on disk |
| Inference VRAM (incl. KV cache @ 32K context) | ~16 GB |
Best at native ≤32K context. For long-context use the AWQ companion via vLLM.
Quick start
from transformers import AutoTokenizer
from hqq.models.hf.base import AutoHQQHFModel
tok = AutoTokenizer.from_pretrained("drawais/Qwen2.5-Coder-14B-Instruct-HQQ-INT4")
model = AutoHQQHFModel.from_quantized("drawais/Qwen2.5-Coder-14B-Instruct-HQQ-INT4", device="cuda")
License
Apache 2.0 (inherits from base model).