// AI Platform Drift

DeepSeek Flash Latest pricing increased 25% completion, 13.6% prompt

OpenRouter model catalogue · published 2026-10-02 21:56 UTC
WORTH KNOWING BUDGET PRICINGMODEL
WHAT CHANGEDOpenRouter raised pricing for deepseek/deepseek-flash-latest effective 2 October 2026. Completion tokens increased from $0.60/M to $0.75/M (+25%). Prompt tokens increased from $0.0264/M to $0.0300/M (+13.6%). A workload generating ten completion tokens for every prompt token will see a 25% total cost increase.
WHY IT MATTERSAny application routing requests through this model identifier will pay 25% more per response starting immediately. If you are on a fixed budget or have cost alerts configured, you will hit thresholds sooner than forecasted.
WHAT TO DOCheck your October invoice for deepseek/deepseek-flash-latest usage. If material, update your cost forecast and notify whoever owns the AI spend budget. If you pinned to this model by name, evaluate whether to switch to a lower-cost alternative or stay and absorb the increase.
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-flash-latest
 name: DeepSeek: DeepSeek Flash Latest
-pricing.prompt: 0.0000000264
-pricing.completion: 0.0000006
+pricing.prompt: 0.00000003
+pricing.completion: 0.00000075
 context_length: 1048576
 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 changes directly affect engineering budget and cost modeling for production workloads.
signalspricing.prompt increased ~13.6% · pricing.completion increased 25% · both price changes on same model in same update
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)