What are the hardware requirements for Qwen2 72B Instruct 4bit?

Qwen2 72B Instruct 4bit requires approximately 40.9 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen2 72B Instruct 4bit and how large is it?

Qwen2 72B Instruct 4bit is developed by mlx-community, a model with 72b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Qwen2 72B Instruct 4bit perform on standard benchmarks?

Qwen2 72B Instruct 4bit has the following published scores: LMSYS ELO 1261. Compare against reference models on this page or on the LLM Explorer leaderboards.

Qwen2 72B Instruct 4bit by mlx-community — VRAM 40.9GB, 32K context

Name: Qwen2 72B Instruct 4bit
Author: mlx-community

Qwen2 72B Instruct 4bit is an open-source language model by mlx-community. Features: 72b LLM, VRAM: 40.9GB, Context: 32K, License: other, Quantized, Instruction-Based, LLM Explorer Score: 0.25, ELO: 1261.

4bit Chat Conversational En Instruct Mlx Quantized Qwen2 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlx-community/Qwen2-72B-Instruct-4bit

Qwen2 72B Instruct 4bit Benchmarks

LMSys ELO 1261 vs 1272 (so35) -0.9%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Qwen2 72B Instruct 4bit Parameters and Internals

Model Type

text-generation

Additional Notes

The model was converted to MLX format using mlx-lm version 0.14.2.

LLM Name	Qwen2 72B Instruct 4bit
Repository 🤗	https://huggingface.co/mlx-community/Qwen2-72B-Instruct-4bit
Base Model(s)	Qwen2 72B Instruct Qwen/Qwen2-72B-Instruct
Model Size	72b
Required VRAM	40.9 GB
Updated	2026-07-14
Maintainer	mlx-community
Model Type	qwen2
Instruction-Based	Yes
Model Files	5.4 GB: 1-of-8 5.3 GB: 2-of-8 5.3 GB: 3-of-8 5.3 GB: 4-of-8 5.3 GB: 5-of-8 5.3 GB: 6-of-8 5.3 GB: 7-of-8 3.7 GB: 8-of-8
Supported Languages	en
Quantization Type	4bit
Model Architecture	Qwen2ForCausalLM
License	other
Context Length	32768
Model Max Length	32768
Transformers Version	4.40.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	152064
Torch Data Type	bfloat16
Errors	replace

Rank the Qwen2 72B Instruct 4bit capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Qwen2 72B Instruct 4bit

Best Alternatives	Context / RAM	Downloads	Likes
...B Instruct 2.0bpw H Novel EXL2	125K / 23 GB	4	1
ECE ILAB Q1	32K / 77.8 GB	15	0
Qwen2.5 72B Instruct Bnb 4bit	32K / 41.4 GB	14954	8
Qwen2.5 72B Instruct 4bit	32K / 40.9 GB	1200	7
Athene V2 Chat 4.65bpw H6 EXL2	32K / 44.3 GB	6	6
Qwen2.5 72B Instruct 8bit	32K / 77.1 GB	614	3
Qwen2 72B Instruct Bpw5.5 EXL2	32K / 51.8 GB	1	1
...struct Mix Calibration 4B EXL2	32K / 38.8 GB	3	2
Qwen2 72B Instruct 8bit	32K / 77.1 GB	48	1
Qwen2 72B Instruct Bpw4 EXL2	32K / 38.7 GB	5	0