// AI Platform Drift

DeepSeek Pro Latest pricing dropped 87% for completions, 17% for prompts

OpenRouter model catalogue · published 2026-10-02 20:04 UTC
WORTH KNOWING BUDGET PRICINGMODEL
WHAT CHANGEDOpenRouter's pricing for deepseek/deepseek-pro-latest changed on 2026-10-01. Prompt pricing decreased from $0.1853/M tokens to $0.1531/M tokens (-17.4%). Completion pricing decreased from $3.5000/M tokens to $0.4594/M tokens (-86.9%).
WHY IT MATTERSIf you are running this model in production, your inference costs just dropped by 83-87% depending on your prompt-to-completion ratio. A workload generating ten completion tokens per prompt token now costs 86.5% less per request.
WHAT TO DOPull your last 30 days of usage for this model from OpenRouter's API or dashboard, apply the new rates to your token counts, and send the revised forecast to whoever owns your 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-pro-latest
 name: DeepSeek: DeepSeek Pro Latest
-pricing.prompt: 0.0000001853
-pricing.completion: 0.0000035
+pricing.prompt: 0.00000015312
+pricing.completion: 0.00000045936
 context_length: 1048576
 top_provider.max_completion_tokens: 393216
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 reasoningSignificant price reductions on a production model affect cost planning and budget forecasts.
signalsprompt pricing decreased ~17% · completion pricing decreased ~87% · pricing change on active model
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)