Qwen2.5-72B-Instruct

Vendor: alibaba
Family: qwen
Tier: small
Status: deprecated
Open weights: yes
Release date: 2024-09-19
Deprecation date: not announced
Context window: 131072
Max output tokens: –
Tokenizer family: qwen2
Parameters: 72B total
Price per intelligence point: Coming soon: benchmark ingestion

Offerings

Host	Variant	Region	Status	Input $/Mtok	Output $/Mtok	Cache read	Cache write	Latency / throughput soon	Confidence	Source
deepinfra	–	–	active	$0.360000	$0.400000	–	–	–	medium	litellm
hyperbolic	–	–	active	$0.120000	$0.300000	–	–	–	medium	litellm
nebius	–	–	active	$0.130000	$0.400000	–	–	–	medium	litellm
together	–	–	active	$1.200000	$1.200000	–	–	–	medium	litellm

Cite this page

Qwen2.5-72B-Instruct: $0.120000/$0.300000 per 1M tokens on hyperbolic as of 2026-10-04. Token Barometer by Dejan Strbad, https://tokenbarometer.com, data CC BY 4.0. https://tokenbarometer.com/model/alibaba/qwen2.5-72b-instruct

Badge

Embed a pricing badge for this model:

[![Qwen2.5-72B-Instruct price](https://tokenbarometer.com/badge/alibaba/qwen2.5-72b-instruct.svg)](https://tokenbarometer.com/model/alibaba/qwen2.5-72b-instruct)

Price history

Change log

Date	Host	Kind	Change
2026-10-04	deepinfra	New offering	New offering
2026-10-04	hyperbolic	New offering	New offering
2026-10-04	nebius	New offering	New offering
2026-10-04	together	New offering	New offering
2026-10-04	together	New offering	New offering

Compare

No comparison pages for this model yet. See all comparisons.