Side-by-side technical and economic comparison between Alibaba / Qwen's Qwen 2.5 72B Instruct — Together AI and DeepSeek's DeepSeek-V4.1-Reasoner.
• Qwen 2.5 72B Instruct — Together AI: Classified as a Legacy Reference model (Deprecated on Together AI (Reference)).
| Metric | Qwen 2.5 72B Instruct — Together AI | DeepSeek-V4.1-Reasoner | Advantage |
|---|---|---|---|
| Provider Organization | Alibaba / Qwen | DeepSeek | — |
| Standard Input / 1M | $0.35 | $0.50 | Qwen 2.5 72B Instruct — Together AI (30% lower) |
| Cached Input / 1M | $0.175 | $0.01 | DeepSeek-V4.1-Reasoner lower |
| Output / 1M | $0.40 | $2.50 | Qwen 2.5 72B Instruct — Together AI lower |
| Context Window | 128k | 256k | DeepSeek-V4.1-Reasoner (256k) |
| Max Generation Tokens | 8.2k | 384k | DeepSeek-V4.1-Reasoner |
| Tokenizer Family | Qwen Byte-level BPE (~152k vocabulary) | DeepSeek Byte-Level BPE (102k vocabulary) | — |
Opt for Qwen 2.5 72B Instruct — Together AI when your engineering requirements prioritize Alibaba / Qwen's ecosystem, specific tokenizer efficiencies (Calibrated Qwen Tokenizer (±3%)), or when your expected prompt-to-completion ratios favor its $0.35/M input rate.
Opt for DeepSeek-V4.1-Reasoner when looking for DeepSeek's tooling integration, specific context window depth (256k tokens), or when output generation volume favors its $2.5/M rate.