Leading open-weights foundation model for deep coding, math, and multilingual reasoning
qwen/qwen-2.5-72b-instruct| Scenario | Input Tokens | Output Tokens | Uncached Cost | With Prompt Caching |
|---|---|---|---|---|
| Short Chat Query | 1,000 | 500 | $0.00055 | $0.00038 |
| Document Summarization | 10,000 | 2,000 | $0.00430 | $0.00255 |
| Codebase & Context Analysis | 100,000 | 20,000 | $0.0430 | $0.0255 |
| Batch Corpus Processing | 1,000,000 | 100,000 | $0.3900 | $0.2150 |
Ideal for production workloads demanding flagship capabilities, deep context depth (128k tokens), and reliability from Alibaba / Qwen. Excellent when predictable tokenomics and prompt caching support are paramount.
If your use-case requires sub-second streaming latency or ultra-high frequency classification at micro-cent pricing, consider lighter budget options such as Gemini Flash-Lite or Claude Haiku. For deep formal logic, consider dedicated reasoning models like o3.
For Qwen 2.5 72B Instruct — Together AI, 1 million input tokens costs $0.35, while 1 million output tokens costs $0.40. If using prompt caching, repetitive input prefixes are discounted to $0.175 per million.
Qwen 2.5 72B Instruct — Together AI features a maximum context window of 131,072 tokens (~98,304 words), with a maximum output limit of 8,192 tokens per completion.
Qwen 2.5 72B Instruct — Together AI utilizes the Qwen Byte-level BPE (~152k vocabulary). Token counting on TokenMath runs client-side to ensure maximum privacy.
Yes. Qwen 2.5 72B Instruct — Together AI supports prompt caching with a cached input rate of $0.175/1M (saving up to 50% on repeated input context).