Updated 17 Sep 2026 • 5 mins read
DeepSeek API Pricing 2026: Models, Token Costs, and How to Optimize
DeepSeek API pricing in 2026 is per million tokens and, since August 2026, split into peak and off-peak rates. This guide covers current V4 Flash and V4 Pro costs, the 1-million-token context window, caching discounts, and how DeepSeek compares with GPT and Claude, with the levers to reduce your bill.
DeepSeek's API is priced per million tokens and, since August 16, 2026, uses a peak and off-peak schedule where off-peak is roughly half price. As of September 2026, DeepSeek V4 Flash costs about $0.22 per million input tokens and $0.66 per million output off-peak (around $0.44 and $1.32 at peak) on a cache miss, and V4 Pro costs about $0.66 and $1.98 off-peak. Cache hits are far cheaper, around $0.007 per million off-peak on Flash, and every V4 model includes a 1-million-token context window. DeepSeek changes its rates frequently, so confirm the current figures on DeepSeek's official pricing page before budgeting.
Key takeaway:
DeepSeek API pricing in 2026 runs on two models. V4 Flash costs $0.14 per million input tokens and $0.28 output, the cheapest frontier-class API available. V4 Pro lists at $1.74/$3.48 with a standing 75% promotional discount that drops it to roughly $0.435/$0.87. Both support a 1M-token context with no long-context surcharge, and cache hits cost about a tenth of the standard input rate.
The DeepSeek Model Lineup in 2026
DeepSeek V4 arrived in April 2026 and replaced the previous lineup of V3.2, R1, and the legacy API aliases. Two models now cover everything, with V4 Flash offering tiered reasoning modes so you only pay for deep reasoning when you need it.
| Model | Input / 1M | Output / 1M | Best for |
|---|---|---|---|
| V4 Flash | $0.14 | $0.28 | General tasks; cheapest frontier-class API |
| V4 Pro (standard) | $1.74 | $3.48 | Hardest reasoning and agentic work |
| V4 Pro (with 75% promo) | $0.435 | $0.87 | Same model at the discounted rate |
V4 Flash supports a Non-Think mode for routine tasks and Think High or Think Max modes for complex reasoning, so a single model spans cheap, fast answers and heavy reasoning. Both V4 Flash and V4 Pro support a 1M-token context window and up to 384k output tokens.
Legacy model migration
If you still call the old aliases, plan to migrate. The legacy deepseek-chat and deepseek-reasoner aliases are scheduled for retirement on July 24, 2026, after which requests return errors. They currently route to V4 Flash non-thinking and thinking modes. Migration is a one-line change to the model parameter (deepseek-v4-flash or deepseek-v4-pro) on the same base URL and API key. Note that deepseek-reasoner maps to Flash, not Pro.
The Free Tier and What It Includes
DeepSeek's consumer chat is genuinely free. Full model access at chat.deepseek.com and in the mobile app costs nothing for individuals, with web search, file uploads, and saved history included, and no Plus or Pro subscription tier at all. The only catch is fair-use throttling, so during peak hours you may see Server Busy warnings.
For developers, every new API account gets a grant of around 5 million free tokens, valid for roughly 30 days, which is enough to prototype before you pay anything. After that it is pure pay-as-you-go with no minimum spend and no monthly fee. This consumption model is exactly the kind of metered AI spend we cover in our token budgeting framework.
The Discount Levers That Cut Your Bill
DeepSeek is already cheap, but two built-in levers cut the bill much further with little effort.
Prompt caching
DeepSeek automatically caches input chunks of 64 tokens or more. Cache hits cost a fraction of cache misses, often around a tenth of the standard input rate, so keeping a stable system prompt or reference content at the start of every request can cut input costs by 80% or more. No code changes are needed beyond structuring the prompt so the prefix stays identical.
Off-peak pricing
DeepSeek has historically applied automatic off-peak discounts during 16:30 to 00:30 UTC, around 50% off the chat model and up to 75% off the reasoner, with no configuration needed. V4 off-peak pricing had not been formally confirmed at the time of writing, so check the official docs before relying on it, but scheduling non-urgent batch work into that window is worth testing.
Stacking the savings: The levers combine. A workload that pins its system prompt for cache hits, routes routine calls to V4 Flash Non-Think mode, and schedules batch jobs into the off-peak window can run at a small fraction of even DeepSeek's already-low list price. The discipline is the same as any token workload: cache hard, route by difficulty, and time-shift what you can.
DeepSeek vs GPT, Claude, and Gemini
DeepSeek's position is simple: frontier-class reasoning at the lowest cost. The comparison below uses representative 2026 rates per million tokens.
| Model | Input / 1M | Output / 1M | Notes |
|---|---|---|---|
| DeepSeek V4 Flash | $0.14 | $0.28 | Cheapest frontier-class option |
| GPT-5 (OpenAI) | $1.25 | $10.00 | Flagship general and reasoning |
| Claude Sonnet 4.6 | $3.00 | $15.00 | Balanced production workhorse |
| Gemini 2.5 Flash-Lite | $0.10 | Low | Cheaper input but smaller, less capable |
Against the major labs, DeepSeek V4 Flash undercuts the frontier tier by an order of magnitude on output while scoring competitively on coding and reasoning benchmarks. For the full picture on the alternatives, see our ChatGPT pricing in 2026, our Claude AI 2026 guide, and our Google Gemini API pricing guides.
DeepSeek API Pricing at a Glance (2026)
Rates are per million tokens in US dollars, on a cache miss, unless noted. Off-peak is roughly half of peak.
| Model | Input (off-peak / peak) | Output (off-peak / peak) | Cache hit input | Context |
|---|---|---|---|---|
| DeepSeek V4 Flash | $0.22 / $0.44 | $0.66 / $1.32 | ~$0.007 / $0.014 | 1M tokens |
| DeepSeek V4 Pro | $0.66 / $1.32 | $1.98 / $3.96 | Separate, lower | 1M tokens |
Peak hours are 01:00 to 04:00 and 06:00 to 10:00 UTC, Monday to Friday; off-peak covers the rest of the week at half price. Figures reflect DeepSeek's official pricing as of September 2026 and change frequently, so verify on the official page before budgeting.
Running DeepSeek Through Third-Party Providers
You can also reach DeepSeek through aggregators and clouds. OpenRouter matches DeepSeek's direct rates for V4 models and adds a free tier for distilled variants. AWS Bedrock and Azure AI Foundry charge a premium but solve data-residency concerns by routing through US and EU infrastructure, which matters for teams that cannot send data to China. Together AI and Fireworks offer competitive rates on Flash-class models but charge more for reasoning models.
For sustained production volume, the direct API generally provides the best cost floor, especially once off-peak and caching are in play. If data residency is your constraint, the hosted routes are worth the premium, the same tradeoff we discuss in our Amazon Bedrock pricing guide.
How to Control DeepSeek API Costs
- Route by difficulty. Use V4 Flash for general tasks and Non-Think mode for routine calls; reserve Think modes and V4 Pro for genuinely hard reasoning.
- Pin prompts for cache hits. Keep system prompts and reference content identical and at the start of each request so the automatic cache fires.
- Time-shift batch work. Schedule non-urgent jobs into the off-peak window where discounts apply, and confirm current off-peak terms in the docs.
- Set max output limits and attribute spend. Cap response length and tag calls by team and feature so you can see cost per outcome, as covered in our LLM cost optimization guide.
Conclusion
DeepSeek API pricing in 2026 remains the value benchmark the rest of the market is measured against. Two models cover the range: V4 Flash at $0.14/$0.28 for the cheapest frontier-class inference available, and V4 Pro for the hardest work, with a standing promotion that keeps it inexpensive. Add automatic caching, off-peak discounts, and a 1M-token context with no surcharge, and the effective cost drops well below even the headline rates. Migrate off the legacy aliases before July 24, route by difficulty, cache hard, and time-shift batch jobs. If you want help attributing and controlling AI and cloud spend across providers, that is exactly the discipline Opslyft brings.
Key Takeaways
- DeepSeek API pricing is per million tokens, split by cache hit versus cache miss and, since August 2026, by peak versus off-peak hours, with off-peak at roughly half price.
- DeepSeek V4 Flash is the low-cost default at about $0.22 input and $0.66 output per million off-peak; V4 Pro is the higher-capability model at about $0.66 and $1.98 off-peak.
- Every V4 model includes a 1-million-token context window and up to 384K output tokens at no surcharge.
- Cache hits cut input cost dramatically, to around $0.007 per million off-peak on Flash, so pinning a stable system prompt saves money.
- The legacy deepseek-chat and deepseek-reasoner model names were retired on July 24, 2026; new code should use the V4 model names.
- DeepSeek is far cheaper than frontier US models on output, which is its core value proposition
FAQs
DeepSeek API pricing centers on two V4 models. V4 Flash costs $0.14 per million input tokens and $0.28 output. V4 Pro lists at $1.74/$3.48 with a standing 75% promotional discount that drops it to about $0.435/$0.87. Both support a 1M-token context.
Yes, significantly. V4 Flash undercuts frontier models from OpenAI and Anthropic by roughly an order of magnitude on output tokens while scoring competitively on many coding and reasoning benchmarks.
As of the April 2026 V4 launch, two models: V4 Flash for general tasks and V4 Pro for the hardest reasoning. They replaced the earlier V3.2 and R1 lineup. V4 Flash also offers Non-Think and Think modes.
Yes. Consumer chat at chat.deepseek.com and the mobile app is free for individuals with web search and file uploads, subject to fair-use throttling. The API gives new accounts about 5 million free tokens for roughly 30 days.
DeepSeek automatically caches input chunks of 64 tokens or more. Cache hits cost roughly a tenth of cache misses, so keeping a stable prompt prefix can cut input costs by 80% or more with no code changes.
DeepSeek caches repeated input prefixes, such as a fixed system prompt or few-shot examples. When a call reuses that cached prefix (a cache hit), the input is billed at a small fraction of the cache-miss rate, around $0.007 per million off-peak on Flash, so keeping your system prompt identical across calls saves money.
Every DeepSeek V4 model includes a 1-million-token context window at no extra charge, with up to 384K output tokens per request. Many competing models charge a premium for extended context, which is part of DeepSeek's cost advantage.
DeepSeek's web chat at chat.deepseek.com is free to use, and new developer accounts have historically received a free API token grant. There is no standing paid consumer subscription; API usage is pay-per-token. Confirm current free-tier terms on DeepSeek's official site.