DeepSeek V4 Pro 0813 vs Gemma 4 31B
API pricing, context, and independently measured performance, side by side.
DeepSeek V4 Pro 0813
DeepSeek · released Aug 12, 2026
Best input$0.10 / MTok
Best output$1.98 / MTok
Context1.05M
Providers25
Fastest measured0.58s° TTFT · 169 tok/s
Open weightsToolsReasoningStructured outputPrompt caching
Gemma 4 31B
Google · released Apr 2, 2026
Best input$0.10 / MTok
Best output$0.34 / MTok
Context262K
Providers19
Fastest measured0.35s° TTFT · 45 tok/s
Open weightsVisionToolsReasoningStructured outputPrompt caching
Cheapest providers, measuredUSD / MTOK
| MODEL | PROVIDER | INPUT | OUTPUT | TTFT | TPS | SMOKE |
|---|---|---|---|---|---|---|
| DeepSeek V4 Pro 0813 |  Relace | $0.10 | $8.00 | 0.69s° | 136° tok/s | — |
| DeepSeek V4 Pro 0813 |  Wafer | $0.117 | $4.66 | 0.58s° | 169° tok/s | — |
| DeepSeek V4 Pro 0813 |  StreamLake | $0.66 | $1.98 | 1.7s° | 61° tok/s | 5/5 |
| Gemma 4 31B |  CoreWeave | $0.10 | $0.34 | 1.6s° | 5° tok/s | 5/5 |
| Gemma 4 31B |  Venice | $0.12 | $0.36 | 1.4s° | 45° tok/s | 5/5 |
| Gemma 4 31B |  Chutes | $0.12 | $0.37 | — | — | 5/5 |
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