What deepseek-v4-pro costs for real workloads
| Workload | Estimated cost |
|---|---|
| Support chatbot: 100K messages | $112.20 |
| RAG search: 10K queries with 6K-token context | $49.50 |
| Batch summarization: 1,000 long documents | $15.18 |
| Coding agent: 100 tasks, 50K context x 20 turns | $9.35 |
Chatbot: 800 input + 300 output tokens per message. RAG: 6,000 + 500 per query. Summarization: 20,000 + 1,000 per document. Coding agent: 20 turns per task over a 50K-token context, with 95% of input served from cache where the model supports it.
deepseek-v4-pro vs similarly priced models
| Model | Provider | Input / 1M | Output / 1M | Cached / 1M | Blended |
|---|---|---|---|---|---|
| Gemini 3 Nano | $0.30 | $2.50 | $0.030 | $0.85 | |
| Gemini 3.5 Flash | $0.30 | $2.50 | $0.030 | $0.85 | |
| Gemini 2 Flash | $0.30 | $2.50 | $0.030 | $0.85 | |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 | $0.030 | $0.85 | |
| deepseek-v4-pro | DeepSeek | $0.66 | $1.98 | $0.022 | $0.99 |
| Gemini 2.5 Flash | $0.075 | $3.75 | $0.075 | $0.99 | |
| gpt-realtime-2.1-mini | OpenAI | $0.60 | $2.40 | $0.060 | $1.05 |
| Mistral Large 4 | Mistral AI | $0.70 | $2.10 | $0.070 | $1.05 |
| Gemini 3 Flash Preview | $0.50 | $3.00 | $0.050 | $1.12 |
Other DeepSeek models
deepseek-v4-pro pricing FAQ
How much does the deepseek-v4-pro API cost?
deepseek-v4-pro costs $0.66 per million input tokens and $1.98 per million output tokens on DeepSeek, with cached input at $0.022 per million. Prices as of November 8, 2026.
How much does deepseek-v4-pro cost for 1,000 requests?
At 1,000 input and 500 output tokens per request, 1,000 requests cost about $1.65.
Does deepseek-v4-pro have a prompt caching discount?
Yes. Cached input is $0.022 per million tokens, 30x cheaper than fresh input. Workloads that resend the same context, like agents and chat with long system prompts, pay mostly the cached rate.
Is deepseek-v4-pro expensive compared to other LLM APIs?
It ranks #31 cheapest of 76 models we track on a blended price (3 input tokens per output token) of $0.99 per million tokens.
Picking a model for production?
Describe your workload and the Planner estimates monthly token spend across every model we track.