Qwen2.5 VL 72B Instruct
alibaba/qwen2.5-vl-72b-instruct- Alibaba
- Family · Qwen2 VL
- Open weights
- Text · Image → Text
- JSON · Vision
Not on OSMEXO yet
Spec & list price
Context125K
Max output115K
Input / 1M$0.80
Output / 1M$1.001.3× input
Usage
Tokens / day145.5M▼ 22% 7d
Usage rank#287
Requests / day77.2K
Tool calls / req0.00
Per request · 30 days
Avg input1.8K0.1× median
Avg output880.1× median
Cache hit25%0.4× median
Tool-call failures99%53× median
Consumption
Peak 451.5M 11 OctLow 145.5M 1 NovRank today #287
News & trends
News · latest from Alibaba
- Memdoor: A team of 5 Qwen3.8-27B models beats Fable 5 on LiveCodeBench Hardgithub.com · 31 Oct
- Qwen3.8-Flash Now Available in the EU on Alibaba Cloudtwitter.com · 31 Oct
- Alibaba's Qwen3.8-Flash-Next model tried to upload a file to Beijing in a testnotebookcheck.net · 27 Oct
- Neuron Explorer for Qwen3.5-4B: See every neuron at work, make targeted editsgithub.com · 26 Oct
Speed
Output speed#285/33432 tok/s2.0× slower than median
First token#1470.94 s1.1× quicker than median
Technical details
- Tokenizer
- Qwen
- Knowledge cutoff
- 2024-06-30
- License
- other
- Parameters
- 73.4B
- Moderation
- none
- Downloads · likes
- 99.1K · 666
Request parameters 14
- response_format
- structured_outputs
- temperature
- top_p
- top_k
- max_tokens
- stop
- seed
- frequency_penalty
- logit_bias
+4 more
- logprobs
- presence_penalty
- repetition_penalty
- top_logprobs
When to use
Best for
- Image understanding in chat apps
- Document and screenshot parsing
- Chart and diagram question answering
- Self-hosted vision pipelines
Think twice for
- Tool-calling agents
- Image generation
- Pure text bulk chat (cheaper options exist)
Want Qwen2.5 VL 72B Instruct on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.