Qwen2.5 VL 72B Instruct
Open weightVisionQwen · Released February 2025 · Knowledge cutoff June 2024
Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.
It is part of Alibaba's Qwen family, a broad catalog where Max leads on capability, Plus balances cost, and Flash and the open-weight sizes cover high-volume work.
Facts and pricing
- Provider
- Qwen
- Context window
- 131K tokens
- Input price
- 8.8 credits / 1M tokens ($0.80 raw)
- Output price
- 11.0 credits / 1M tokens ($1.00 raw)
- Vision (image input)
- Yes
- Tool calling
- No
- Extended reasoning
- No
Credits are what Zeplik bills: 1 credit = $0.10, computed from the raw provider rate with a 1.10x margin. Raw prices shown per 1M tokens.
Try Qwen2.5 VL 72B Instruct now
What Qwen2.5 VL 72B Instruct is best for
- Everyday chat and drafting on an open-weight model with transparent lineage
- Working with images: screenshots, charts, photos and scanned documents alongside text
Example prompts
Prompts that suit a open weight model like Qwen2.5 VL 72B Instruct:
- Draft a project update from these rough notes
- Explain how HTTPS works to a curious teenager
- Turn this list of features into a changelog entry
- Brainstorm objections to this proposal and how to answer them
Qwen family
| Model | Context | Input cr/M | Output cr/M | Released |
|---|---|---|---|---|
| Qwen3.7 Plus | 1M | 3.5 | 14.1 | June 2026 |
| Qwen3.7 Max | 1M | 13.8 | 41.3 | May 2026 |
| Qwen3.5 Plus 2026-04-20 | 1M | 3.3 | 19.8 | April 2026 |
| Qwen3.6 Flash | 1M | 2.1 | 12.4 | April 2026 |
| Qwen3.6 35B A3B | 262K | 1.5 | 11.0 | April 2026 |
| Qwen3.6 Max Preview | 262K | 11.4 | 68.6 | April 2026 |
| Qwen3.6 27B | 262K | 3.1 | 26.4 | April 2026 |
| Qwen3.6 Plus | 1M | 3.6 | 21.4 | April 2026 |
| Qwen3.5-9B | 262K | 1.1 | 1.7 | March 2026 |
| Qwen3.5-35B-A3B | 262K | 1.5 | 11.0 | February 2026 |
Frequently asked questions
- What is Qwen2.5 VL 72B Instruct?
- Qwen2.5 VL 72B Instruct is an AI model by Qwen. Qwen2.5-VL is proficient in recognizing common objects such as flowers, birds, fish, and insects. It is also highly capable of analyzing texts, charts, icons, graphics, and layouts within images.
- How much does Qwen2.5 VL 72B Instruct cost on Zeplik?
- Input tokens cost 8.8 credits per million and output tokens 11.0 credits per million (1 credit = $0.10; the raw provider rates are $0.80 and $1.00 per million). New accounts start with free credits.
- How long can a conversation with Qwen2.5 VL 72B Instruct be?
- Qwen2.5 VL 72B Instruct has a 131K-token context window (131,072 tokens), which covers the conversation plus any documents you attach.
- Does Qwen2.5 VL 72B Instruct support images and tools?
- Qwen2.5 VL 72B Instruct accepts image input and does not support tool calling.
Related models
Qwen3.7-Plus is a cost-effective model in Alibaba's Qwen3.7 series. It supports text and image input with text output, building on the series' text capabilities with a comprehensive upgrade to its...
Qwen3.7-Max is the flagship model in Alibaba's Qwen3.7 series. It supports text input and output and is designed for agent-centric workloads, with particular strengths in coding, office and productivity tasks,...
Qwen3.5 Plus (April 2026) is a large-scale multimodal language model from Alibaba. It accepts text, image, and video input and produces text output, with a 1M token context window. This...
Qwen3.6 Flash is a fast, efficient language model from Alibaba's Qwen 3.6 series. It supports text, image, and video input with a 1M token context window. Tiered pricing kicks in...
Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token. It uses a hybrid sparse mixture-of-experts architecture combining Gated...
Qwen3.6-Max-Preview is a proprietary frontier model from Alibaba Cloud built on a sparse mixture-of-experts architecture with approximately 1 trillion total parameters. It is optimized for agentic coding, tool use, and...