Alibaba/Dense

Qwen2.5-72B

Name: Qwen2.5-72B
Author: Alibaba

chattool_use

72.7B

Parameters

32K

Context length

Benchmarks

Quantizations

755K

HF downloads

Architecture

Dense

Released

2024-09-16

Layers

KV Heads

Head Dim

128

Family

qwen

Quantization Options

Quant	Bits	VRAM	Quality
IQ2_M	2.93	27.1 GB	low
Q2_K	3.16	29.2 GB	low
IQ3_XXS	3.25	30.0 GB	low
IQ3_XS	3.5	32.3 GB	low
Q3_K_S	3.64	33.6 GB	low
IQ3_M	3.76	34.7 GB	low
Q3_K_M	4	36.8 GB	low
Q3_K_L	4.3	39.6 GB	moderate
IQ4_XS	4.46	41.0 GB	moderate
Q4_K_S	4.67	42.9 GB	moderate
Q4_K_M	4.89	44.9 GB	good
Q5_K_S	5.57	51.1 GB	good
Q5_K_M	5.7	52.3 GB	good
Q6_K	6.56	60.1 GB	excellent
Q8_0	8.5	77.7 GB	lossless
FP16	16	145.9 GB	lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN QWEN2.5-72B NOW

Rent on RunPod →Or Vast.ai →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (15)

Arena Elo1481

IFEval86.4

BBH61.9

MBPP61.6

MATH59.8

HumanEval59.1

MMLU-PRO51.4

GPQA Diamond49.1

BigCodeBench38.5

LiveCodeBench27.6

GPQA16.7

AIME14.0

MATH-50014.0

MUSR11.7

HLE4.2

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run qwen2.5:72.7b-instruct-q4_K_M

Downloads and runs automatically. Add --verbose for speed stats.

▸ SETUP GUIDE

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more

Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

HuggingFace Ollama Library GGUF Downloads Build Hardware