Qwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. Usage \| API reference \| Try online ## International If you select the International deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.
function_callinghttps://dashscope-intl.aliyuncs.com/compatible-mode/chat| Provider | Input | Output | Since launch |
|---|---|---|---|
| Alibaba Cloudcheapest | $0.18 | $0.7 | — |
| Qwencheapest | $0.18 | $0.7 | — |
| Model | Context | Max out | Pricing |
|---|---|---|---|
| qwen2.5-72b-instruct-1m | 1M | — | |
| qwen2.5-omni-7b | 1M | 2K | |
| qwen2.5-vl-32b-instruct | 1M | — | |
| qwen2.5-vl-3b-instruct | 1M | — | |
| qwen2.5-vl-72b-instruct | 1M | — | |
| qwen2.5-vl-7b-instruct | 1M | — | |
| qwen2.5-math-7b-instruct | — | — | $0.14$0.29 |
| qwen2.5-7b-instruct | 1M | — | $0.18$0.7 |
| qwen2.5-14b-instruct | 1M | — | $0.35$1.4 |
| qwen2.5-7b-instruct-1m | 1M | — | $0.37$1.47 |
| qwen2.5-math-72b-instruct | 4K | 3K | $0.57$1.72 |
| qwen2.5-32b-instruct | 1M | — | $0.7$2.8 |
| qwen2.5-14b-instruct-1m | 1M | 8K | $0.81$3.22 |
| qwen2.5-72b-instruct | 1M | — | $1.4$5.6 |
/v1/models/qwen/qwen2.5-7b-instructQwen2.5 is part of the Qwen Large Language Model (LLM) series. We released a series of base language models and instruction-tuned language models with parameter scales ranging from 7 billion to 72 billion. Qwen2.5 improves upon Qwen2 in the following ways: * Pre-trained on our latest large-scale dataset containing up to 18 trillion tokens. * Significantly expanded knowledge and greatly enhanced coding and math capabilities due to our domain-expert models. * Major improvements in instruction following, long-text generation (over 8K tokens), structured data understanding (such as tables), and structured output generation (especially JSON). More resilient to diverse system prompts, enhancing chatbot role-playing and conditional setup. * Supports over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, and Arabic. Usage \| API reference \| Try online ## International If you select the International deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.