Qwen 2.5 72B

Alibaba / Qwen
Text & Chat

Alibaba's powerful open-source model. Excellent at coding, math, and multilingual tasks.

Try Qwen 2.5 72B now
Send a single prompt and stream a response inline. Hit Cmd+Enter to submit.
Sign in to try this model with โ‚ฌ5 free credits.
Sign in
Press Cmd+Enter to send
Response appears here.
TL;DRยทLast updated March 4, 2026

Qwen 2.5 72B is text & chat AI model from Alibaba / Qwen, priced at โ‚ฌ12.00 per 1M input tokens with a 131.1K tokens context window.

Try Qwen 2.5 72B

0.7

Sign in to generate โ€” 50 free credits on sign-up

Examples

See what Qwen 2.5 72B can generate

Data Analysis

Given monthly sales data [Jan: 12000, Feb: 15000, Mar: 14500, Apr: 18000, May: 22000, Jun: 19500], identify the trend and forecast July

The data shows a general upward trend with month-over-month growth averaging about 12.5%. There's a notable spike in May (+22.2%) followed by a correction in June (-11.4%), suggesting possible seasonality or a promotional event in May. Using linear regression on the six data points, the projected July sales are approximately 21,200. However, given the May-June pullback pattern, a more conservative estimate of 20,000-21,500 accounts for potential volatility.

Regex Help

Write a regex pattern that validates email addresses and explain each part

Pattern: `^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}$` Breakdown: - `^` โ€“ Start of string - `[a-zA-Z0-9._%+-]+` โ€“ One or more valid local-part characters (letters, digits, dots, underscores, percent, plus, hyphen) - `@` โ€“ Literal @ symbol - `[a-zA-Z0-9.-]+` โ€“ Domain name (letters, digits, dots, hyphens) - `\.` โ€“ Literal dot before TLD - `[a-zA-Z]{2,}` โ€“ TLD must be at least 2 letters - `$` โ€“ End of string Note: This covers most common emails but won't match every valid RFC 5322 address.

Pricing

Price per Generation
Per generationFree

API Integration

Use our OpenAI-compatible API to integrate Qwen 2.5 72B into your application.

Install
npm install railwail
JavaScript / TypeScript
import railwail from "railwail";

const rw = railwail("YOUR_API_KEY");

// Simple โ€” just pass a string
const reply = await rw.run("qwen-2-5-72b", "Hello! What can you do?");
console.log(reply);

// With message history
const reply2 = await rw.run("qwen-2-5-72b", [
  { role: "system", content: "You are a helpful assistant." },
  { role: "user", content: "Explain quantum computing simply." },
]);
console.log(reply2);

// Full response with usage info
const res = await rw.chat("qwen-2-5-72b", [
  { role: "user", content: "Hello!" },
], { temperature: 0.7, max_tokens: 500 });
console.log(res.choices[0].message.content);
console.log(res.usage);
Specifications
Context window
131,072 tokens
Max output
4,096 tokens
Avg. latency
2.5s
Developer
Alibaba / Qwen
Category
Text & Chat
Tags
open-source
coding
multilingual

Deep dive โ€” Alibaba Cloud (Qwen team)'s Qwen 2.5 72B

About Alibaba Cloud (Qwen team)
Founded 2009 ยท Hangzhou, China

The Qwen (้€šไน‰ๅƒ้—ฎ, Tongyi Qianwen) team is the LLM research group inside Alibaba Cloud, the cloud-computing arm of Alibaba Group founded in 2009. Alibaba started large-language-model research at the Damo Academy and Tongyi Lab; the Qwen series became publicly available in 2023. Releases include Qwen-7B (Aug 2023, first Chinese open-weight foundation model from a major cloud), Qwen-14B/72B (late 2023), Qwen 1.5 (Feb 2024), Qwen2 family (Jun 2024), Qwen2.5 family (Sep 2024 with 0.5B/1.5B/3B/7B/14B/32B/72B sizes plus Qwen2.5-Coder, Qwen2.5-Math, Qwen2.5-VL), Qwen3 family in 2025 introducing thinking mode, and Qwen2.5-Max as the closed-weight flagship. The Qwen team is led by Junyang Lin and has published over a dozen widely cited technical reports. Models are released under the Tongyi Qianwen LICENSE (commercial-friendly for most use cases but with a >100M MAU clause requiring a separate license). Qwen has become the dominant open-weight model family for Chinese-language tasks and has the largest derivative-model ecosystem on HuggingFace by download volume.

Visit Alibaba Cloud (Qwen team) โ†’
Architecture
Decoder-only Transformer (dense, Grouped Query Attention)

Qwen2.5-72B was released by the Alibaba Qwen team in September 2024 as the flagship dense model of the Qwen2.5 family. It is a decoder-only Transformer with 72.7B parameters, 80 layers and Grouped Query Attention (GQA) with 64 query heads and 8 KV heads. The model was pretrained on a 18-trillion-token multilingual corpus emphasising Chinese, English and 27 other languages, plus heavy concentrations of math, code and long-form documents. Compared to Qwen2-72B (7T tokens), the 18T-token pretraining and improved data filtering pipeline meaningfully lifted knowledge, math and code performance. Qwen2.5-72B supports a 128K token (131,072) context window with YaRN scaling extension and ships in both base and Instruct variants. Post-training applied a multi-stage SFT pipeline on over 1M curated examples followed by Direct Preference Optimisation (DPO) and Group Relative Policy Optimisation (GRPO) for reasoning data. Qwen2.5-Coder-32B and Qwen2.5-Math-72B specialised siblings push code and math frontiers, and Qwen2.5-VL adds vision. Weights are released under the Tongyi Qianwen LICENSE (free commercial use unless >100M MAU). The Qwen ecosystem also publishes GGUF, AWQ, GPTQ quantised variants and is supported by vLLM, SGLang, llama.cpp, Ollama, MLX and Apple MLX.

Parameters
72.7B (dense)
Context
131.1K tokens
What it can do
  • 72.7B dense parameters with Grouped Query Attention
  • Pretrained on 18T multilingual tokens
  • 128K context window with YaRN scaling
  • Strong multilingual performance: Chinese, English, Japanese, Korean and 27 more
  • Specialised siblings: Qwen2.5-Coder, Qwen2.5-Math, Qwen2.5-VL
  • Function calling and JSON mode supported
  • Open weights under Tongyi Qianwen LICENSE (commercial-friendly)
  • Massive ecosystem: vLLM, SGLang, llama.cpp, Ollama, MLX, HuggingFace
  • GGUF/AWQ/GPTQ quantised variants officially published
  • Strong math performance (Qwen2.5-Math variant beats GPT-4o on competition math)
  • Best for: open-weight Chinese/English chat, coding, on-prem enterprise, RAG.
Training & License

Pretrained on 18 trillion tokens of multilingual web text, code, books and scientific papers, with strong Chinese and English coverage. Post-training uses 1M+ curated SFT examples followed by DPO and GRPO. Data cutoff approximately mid-2024.

License: Tongyi Qianwen LICENSE: free commercial use for products with fewer than 100M monthly active users. >100M MAU requires a separate commercial license from Alibaba.

Known limitations
  • Filters Chinese political topics
  • Tongyi Qianwen LICENSE adds a >100M MAU clause
  • Vision requires the separate Qwen2.5-VL checkpoint
  • Knowledge cutoff mid-2024
  • Long context recall quality degrades beyond ~64K

Frequently asked questions

Start using Qwen 2.5 72B today

Get started with free credits. No credit card required. Access Qwen 2.5 72B and 100+ other models through a single API.