DeepSeek V4 Flash 0731 vs Kimi K3
API pricing, context, and independently measured performance, side by side. DeepSeek V4 Flash 0731 starts 98% cheaper on input than Kimi K3 at list prices.
DeepSeek V4 Flash 0731
DeepSeek · released Jul 31, 2026
Best input$0.015 / MTok
Best output$0.132 / MTok
Context1.05M
Providers30
Fastest measured0.38s° TTFT · 265 tok/s
Open weightsToolsReasoningStructured outputPrompt caching
Kimi K3
Moonshot AI · released Jul 16, 2026
Best input$0.69 / MTok
Best output$11.25 / MTok
Context1.05M
Providers25
Fastest measured0.65s° TTFT · 100 tok/s
Open weightsVisionToolsReasoningStructured outputPrompt caching
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| DeepSeek V4 Flash 0731 |  Relace | $0.015 | $1.28 | 0.69s° | 102° tok/s | 5/5 |
| DeepSeek V4 Flash 0731 | OPOpenInference | $0.015 | $1.28 | 1.9s° | 33° tok/s | 3/5 |
| DeepSeek V4 Flash 0731 | SRSail Research | $0.019 | $0.30 | 1.6s° | 50° tok/s | 5/5 |
| Kimi K3 | Relace | $0.69 | $13.00 | 1.1s° | 150° tok/s | — |
| Kimi K3 |  Wafer | $0.70 | $16.00 | 0.87s° | 69° tok/s | 0/5 |
| Kimi K3 |  InferenceNet | $0.72 | $13.00 | 57.9s° | 7° 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