// AI Platform Drift

DeepSeek Pro Latest completion pricing dropped 95%

OpenRouter model catalogue · published 2026-10-17 10:48 UTC
WORTH KNOWING BUDGET PRICINGMODEL
WHAT CHANGEDOpenRouter's ~deepseek/deepseek-pro-latest model completion rate decreased from $8.00 to $0.3831 per 1M tokens, and prompt rate decreased from $0.234 to $0.12332 per 1M tokens, effective as of 17 October 2026.
WHY IT MATTERSUsing the captured base rates for ~deepseek/deepseek-pro-latest, 100M completion tokens per day cost $800.00 before and $38.31 after, excluding all other charges. The 95% completion cost reduction makes this model substantially cheaper for generation-heavy workloads.
WHAT TO DOReforecast your OpenRouter spend if you use this model or are evaluating alternatives. Using the captured base rates for ~deepseek/deepseek-pro-latest, 500K completion tokens per request cost $4.00 before and $0.19155 after, excluding all other charges.
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: 0.000000234
-pricing.completion: 0.000008
+pricing.prompt: 0.00000012332
+pricing.completion: 0.0000003831
 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 reasoningPricing for a production model decreased substantially on both prompt and completion tokens.
signalspricing.prompt decreased from 0.000000234 to 0.00000012332 USD/token · pricing.completion decreased from 0.000008 to 0.0000003831 USD/token · both price fields changed significantly downward
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)