// AI Platform Drift

DeepSeek V4 Flash pricing split: prompt down 37%, completion up 33%

OpenRouter model catalogue · published 2026-09-29 04:41 UTC
WORTH KNOWING BUDGET PRICINGMODEL
WHAT CHANGEDOpenRouter changed pricing for deepseek/deepseek-v4-flash-latest effective 29 September 2026. Prompt tokens dropped from $0.0190/M to $0.0120/M (-36.8%). Completion tokens rose from $0.3000/M to $0.4000/M (+33.3%).
WHY IT MATTERSNet cost impact depends on your completion-to-prompt ratio. Workloads generating three completion tokens per prompt token see a 31.9% total cost increase. Short-output use cases (summaries, classifications) benefit; long-output use cases (generation, code) cost more.
WHAT TO DOCalculate your typical completion-to-prompt ratio from last month's logs. If you generate more than one completion token per prompt token, reforecast your DeepSeek spend upward by 29-33% and notify whoever owns the AI budget.
Model-reported confidence 95% (not measured accuracy) Primary source ↗ written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
 id: ~deepseek/deepseek-v4-flash-latest
 name: DeepSeek: DeepSeek V4 Flash Latest
-pricing.prompt: 0.000000019
-pricing.completion: 0.0000003
+pricing.prompt: 0.000000012
+pricing.completion: 0.0000004
 context_length: 1310720
 top_provider.max_completion_tokens: 943718
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
triage verdictMATERIAL at 0.95, needed 0.5
triage reasoningPricing per token changed on a production model. Affects cost calculations for existing and new deployments.
signalsprompt pricing decreased · completion pricing increased · pricing change on active model
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)