DeepSeek API Cost Calculator

DeepSeek V4.1 Flash API Pricing Explained

DeepSeek V4.1 Flash uses time-based pricing. Off-peak rates per 1M tokens are $0.003 for a cache hit, $0.15 for a cache miss/input, and $0.60 for output. Peak rates are double: $0.006, $0.30, and $1.20. Peak windows are Monday to Friday, 01:00 to 04:00 UTC and 06:00 to 10:00 UTC. Choose the pricing period and cache status above instead of budgeting from the cheapest rate alone.

Correction, September 14: current DeepSeek documentation lists deepseek-v4-pro and deepseek-v4-flash as separate production models. Do not plan for an automatic Pro-to-Flash migration based on the earlier cutoff announcement.

When to Use Each DeepSeek Model

• V4.1 Flash ($0.15/$0.60 off-peak cache-miss input/output): Current production choice for multimodal and high-volume workloads. Cache-hit input is $0.003/M off-peak.
• V4 Pro: Available at its existing price until the September 14 routing cutoff. Do not start a new integration assuming this endpoint will keep serving V4 Pro.

DeepSeek vs Competitors

DeepSeek's time-based rates can be attractive for flexible workloads, but model economics depend on cache hits, output volume, and when requests run. Compare the current schedules with the Gemini 3.8 Flash pricing guide and GPT-6 Astra API pricing calculator before choosing on headline input price alone.

How to Reduce Your DeepSeek API Costs

• Schedule flexible requests off-peak: V4.1 Flash peak rates are twice the off-peak rates, so time-shifting eligible batch work can reduce token charges.
• Preserve cacheable prefixes: Off-peak cache-hit input costs $0.003/M compared with $0.15/M for cache-miss input.
• Leverage the 1M context window: Include all relevant context in a single request instead of making multiple smaller calls.
• Set token limits: Control output length with max_tokens to avoid surprise costs on verbose responses.
• Batch requests: Process multiple items in a single prompt to reduce per-request overhead.
• Monitor usage: Track your API usage to identify cost spikes and optimize patterns.

DeepSeek Free Tier

DeepSeek offers $5 in free credits for new accounts. This is enough for approximately 11.4M input tokens on V4 Pro or 35.7M input tokens on V4 Flash. Great for prototyping and evaluation.

Related Tools

• GPT-5 API Cost Calculator — Compare OpenAI pricing
• Claude API Cost Calculator — Compare Anthropic pricing
• Gemini API Cost Calculator — Compare Google pricing
• xAI Grok Cost Calculator — Compare xAI pricing
• DeepSeek V4 Pricing Guide — Full pricing breakdown
• Cost Optimizer — Get a personalized optimization report