DeepSeek V3.2 vs Kimi K3
API pricing, context, and independently measured performance, side by side. DeepSeek V3.2 starts 92% cheaper on input than Kimi K3 at list prices.
DeepSeek V3.2
DeepSeek · released Dec 1, 2025
Best input$0.209 / MTok
Best output$0.31 / MTok
Context164K
Providers14
Fastest measured0.7s° TTFT · 102 tok/s
Open weightsToolsReasoningStructured outputPrompt caching
Kimi K3
Moonshot AI · released Jul 16, 2026
Best input$2.60 / MTok
Best output$13.00 / MTok
Context1.05M
Providers15
Fastest measured0.61s° TTFT · 84 tok/s
Open weightsVisionToolsReasoningStructured outputPrompt caching
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| DeepSeek V3.2 | GMIGMICloud | $0.209 | $0.31 | 4.1s° | 33° tok/s | 5/5 |
| DeepSeek V3.2 | STStreamLake | $0.214 | $0.322 | 1.8s° | 28° tok/s | 5/5 |
| DeepSeek V3.2 | DIDigitalOcean | $0.25 | $0.80 | 0.7s° | 120° tok/s | 5/5 |
| Kimi K3 | SRSail Research | $2.60 | $13.00 | 1.5s° | 144° tok/s | 2/5 |
| Kimi K3 | MOMorph | $2.80 | $14.00 | 1.7s° | 17° tok/s | 1/5 |
| Kimi K3 |  DeepInfra | $2.85 | $14.25 | 2s° | 25° tok/s | 4/5 |
TTFT/TPS: ModelIndex-measured medians (P50, trailing 72h) · 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