DeepSeek · Updated 2026-07-22

DeepSeek-V3 API Pricing 2026: $0.27/M Tokens

Complete per-million-token pricing for DeepSeek-V3 from DeepSeek, including cached and batch rates, cost-per-1K/1M/10M breakdowns, a monthly cost calculator, and a head-to-head comparison with GPT-4o-mini, Gemini 2.5 Flash, Claude Haiku 3.5. For full cross-provider rankings, open the main calculator.

DeepSeek-V3 headline pricing

Input

$0.27 /M tokens

Output

$1.10 /M tokens

Context · capabilities

128K context · Chat, Code · benchmark 8/10

DeepSeek-V3 cost per 1K, 1M and 10M tokens

Standard on-demand USD rates. Input assumes prompt tokens; output assumes generated tokens.

VolumeInput costOutput costTotal (1:1)
1K tokens$0.0003$0.0011$0.0014
1M tokens$0.27$1.10$1.37
10M tokens$2.70$11.00$13.70

DeepSeek-V3 monthly cost calculator

1M
500K

$0.82

per month for 1M in + 500K out

$9.84

per year

Need to weigh it against other models? Use the full cross-provider calculator.

DeepSeek-V3 vs GPT-4o-mini vs Gemini 2.5 Flash vs Claude Haiku 3.5

DeepSeek-V3 ranks #3 of 4 by blended per-million-token price in this comparison. Lower input + output is cheaper.

ModelInput $/MOutput $/MIn+OutContext
GPT-4o-miniOpenAI$0.15$0.60$0.75128K
Gemini 2.5 FlashGoogle$0.15$0.60$0.751M
YOUDeepSeek-V3DeepSeek$0.27$1.10$1.37128K
Claude Haiku 3.5Anthropic$0.80$4.00$4.80200K

Related pricing pages

DeepSeek-V3 pricing FAQ

How much does DeepSeek-V3 cost?

DeepSeek-V3 is priced at $0.27 per million input tokens and $1.10 per million output tokens on DeepSeek. At a typical 2:1 input-to-output ratio, that works out to roughly $0.5467 per million blended tokens. Context window is 128K.

Is DeepSeek-V3 cheaper than GPT-4o-mini?

No — GPT-4o-mini is cheaper. Its blended input+output rate ($0.75/M) undercuts DeepSeek-V3 ($1.37/M) by about 45%. DeepSeek-V3 may still win on quality, reasoning depth, or ecosystem fit for workloads where that matters more than raw price.

Does DeepSeek-V3 support prompt caching?

No. DeepSeek-V3 does not currently publish a discounted cached-input rate, so all input is billed at the standard $0.27/M. If your workload is cache-heavy, consider GPT-4o-mini or another model that supports prompt caching.

Can I use DeepSeek-V3 with the Batch API?

No published batch discount is available for DeepSeek-V3. For non-latency-sensitive workloads, compare against GPT-4o-mini or another model that offers a batch tier.

☕ Support this tool