DeepSeek V4 Pro (0813) vs Qwen3.8 27B
Price and specs side by side · USD per 1M tokens · verified 2026-11-21
// side by side
| metric | DeepSeek V4 Pro (0813) | Qwen3.8 27B |
|---|
| input / 1M | $0.66 | $0.20 |
| output / 1M | $1.98 | $2.50 |
| blended (3:1) | $0.99 | $0.775 |
| cache read / 1M | $0.022 | $0.05 |
| context | 1.05M | 262.1K |
| max output | 393.2K | 32.8K |
| input types | text | text · image · video |
| reasoning | yes | yes |
| tool calling | yes | yes |
| structured output | yes | yes |
| open weights | yes | yes |
| released | 2026-08-13 | 2026-08-14 |
| knowledge cutoff | n/a | n/a |
| lab | DeepSeek | Alibaba (Qwen) |
// monthly cost scenarios
| workload | tokens in / out | DeepSeek V4 Pro (0813) | Qwen3.8 27B |
|---|
| Chatbot | 10M / 2M | $10.56 | $7.00 |
| RAG app | 50M / 5M | $42.90 | $22.50 |
| Coding agent | 20M / 10M | $33.00 | $29.00 |
// verdict
- Qwen3.8 27B is 21.7% cheaper than DeepSeek V4 Pro (0813) on a blended 3:1 input/output basis.
- DeepSeek V4 Pro (0813) has the larger context window (1.05M vs 262.1K tokens).
- Only Qwen3.8 27B accepts image input.
- Chatbot (10M input / 2M output tokens per month): Qwen3.8 27B costs $7.00 vs $10.56 for DeepSeek V4 Pro (0813), saving $3.56 per month.
- RAG app (50M input / 5M output tokens per month): Qwen3.8 27B costs $22.50 vs $42.90 for DeepSeek V4 Pro (0813), saving $20.40 per month.
- Coding agent (20M input / 10M output tokens per month): Qwen3.8 27B costs $29.00 vs $33.00 for DeepSeek V4 Pro (0813), saving $4.00 per month.
// DeepSeek V4 Pro (0813) or Qwen3.8 27B: which one to choose
- Choose Qwen3.8 27B for cost-sensitive, high-volume workloads.
- Choose DeepSeek V4 Pro (0813) for long documents and large codebases (1.05M-token context).
- Choose Qwen3.8 27B if you need image input.