DeepSeek V4 Pro 0813 vs Qwen3.8 27B
API pricing, context, and independently measured performance, side by side. Qwen3.8 27B starts 81% cheaper on input than DeepSeek V4 Pro 0813 at list prices.
DeepSeek V4 Pro 0813
DeepSeek · released Aug 12, 2026
Best input$0.132 / MTok
Best output$0.396 / MTok
Context1.05M
Providers23
Fastest measured0.57s° TTFT · 116 tok/s
Open weightsToolsReasoningStructured outputPrompt caching
Qwen3.8 27B
Qwen · released Aug 14, 2026
Best input$0.025 / MTok
Best output$1.49 / MTok
Context1M
Providers19
Fastest measured0.91s° TTFT · 127 tok/s
Open weightsVisionToolsReasoningStructured outputPrompt caching
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| DeepSeek V4 Pro 0813 |  Baidu | $0.132 | $0.396 | 2s° | 53° tok/s | — |
| DeepSeek V4 Pro 0813 |  Wafer | $0.14 | $4.20 | — | — | — |
| DeepSeek V4 Pro 0813 |  Ionstream | $0.213 | $1.74 | 1.9s° | 27° tok/s | — |
| Qwen3.8 27B | Wafer | $0.025 | $4.35 | — | — | — |
| Qwen3.8 27B |  Reka | $0.025 | $4.35 | 1s° | 48° tok/s | 4/5 |
| Qwen3.8 27B | DEDekaLLM | $0.049 | $3.00 | 1.5s° | 263° tok/s | — |
TTFT/TPS: ModelIndex-measured medians (P50, trailing 21d) · chat10k profile (~10K in / 200 out) · ° = measured via OpenRouter routing (adds a hop) · SMOKE: of 5 programmatic capability probes passed · methodology v1.1 — full details on each model page