LLM EXPLORER 60,645 MODELS INDEXED

Qwen2.5 72B Instruct GPTQ Int8 by Qwen

By Qwen · 1893 downloads
HF Repo ↗ Copy URL

Qwen2.5 72B Instruct GPTQ Int8 is an open-source language model by Qwen. Features: 72b LLM, VRAM: 77GB, Context: 32K, License: other, Quantized, Instruction-Based, LLM Explorer Score: 0.16.

  Arxiv:2309.00071   Arxiv:2407.10671   8-bit Base model:quantized:qwen/qwen... Base model:qwen/qwen2.5-72b-in...   Chat   Conversational   En   Endpoints compatible   Gptq   Instruct   Quantized   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen2.5-72B-Instruct-GPTQ-Int8 

Qwen2.5 72B Instruct GPTQ Int8 Parameters and Internals

Model Type 
Causal Language Models
Additional Notes 
This model is instruction-tuned, with enhancements in various areas, including long-context support and multilingual capabilities.
Supported Languages 
en (proficient), cn (proficient), fr (proficient), es (proficient), pt (proficient), de (proficient), it (proficient), ru (proficient), ja (proficient), ko (proficient), vi (proficient), th (proficient), ar (proficient)
Training Details 
Context Length:
131072
Model Architecture:
transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
Input Output 
Accepted Modalities:
text
Performance Tips:
To handle extensive inputs exceeding 32,768 tokens, utilize YaRN.
LLM NameQwen2.5 72B Instruct GPTQ Int8
Repository 🤗https://huggingface.co/Qwen/Qwen2.5-72B-Instruct-GPTQ-Int8 
Base Model(s)  Qwen/Qwen2.5-72B-Instruct   Qwen/Qwen2.5-72B-Instruct
Model Size72b
Required VRAM77 GB
Updated2026-06-28
MaintainerQwen
Model Typeqwen2
Instruction-BasedYes
Model Files  3.8 GB: 1-of-20   3.9 GB: 2-of-20   3.9 GB: 3-of-20   3.9 GB: 4-of-20   3.9 GB: 5-of-20   3.9 GB: 6-of-20   3.9 GB: 7-of-20   3.9 GB: 8-of-20   3.9 GB: 9-of-20   3.9 GB: 10-of-20   3.9 GB: 11-of-20   3.9 GB: 12-of-20   3.9 GB: 13-of-20   3.9 GB: 14-of-20   3.9 GB: 15-of-20   3.9 GB: 16-of-20   3.9 GB: 17-of-20   3.9 GB: 18-of-20   3.9 GB: 19-of-20   3.0 GB: 20-of-20
Supported Languagesen
GPTQ QuantizationYes
Quantization Typegptq
Model ArchitectureQwen2ForCausalLM
Licenseother
Context Length32768
Model Max Length32768
Transformers Version4.39.3
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typefloat16
Errorsreplace

Best Alternatives to Qwen2.5 72B Instruct GPTQ Int8

Best Alternatives
Context / RAM
Downloads
Likes
Qwen2.5 72B Instruct GPTQ Int432K / 41.6 GB4839944
Qwen2 72B Instruct GPTQ Int832K / 77 GB170215
Qwen2 72B Instruct GPTQ Int432K / 41.6 GB9633
...B Instruct 2.0bpw H Novel EXL2125K / 23 GB41
Qwen2 72B Instruct 4bit32K / 40.9 GB562
ECE ILAB Q132K / 77.8 GB150
Qwen2.5 72B Instruct Bnb 4bit32K / 41.4 GB149548
Qwen2.5 72B Instruct 4bit32K / 40.9 GB12007
Athene V2 Chat 4.65bpw H6 EXL232K / 44.3 GB66
Qwen2.5 72B Instruct 8bit32K / 77.1 GB6143
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen2.5-72B-Instruct-GPTQ-Int8.