| Input / 1M tokens | Output / 1M tokens | Relative Cost | |
|---|---|---|---|
| DeepSeek V4 Flash | $0.14 | $0.28 | 1x (baseline) |
| DeepSeek V4 Pro | $0.435 | $0.87 | ~3x more |
| Gemini 2.5 Pro | $1.25 | $10.00 | ~9x to 36x more |
| Claude Sonnet 4.6 | $3.00 | $15.00 | ~21x to 54x more |
| Claude Opus 4.6 | $5.00 | $25.00 | ~36x to 89x more |
$0
V4 Flash input per 1M tokens
$0
V4 Flash cache hit input per 1M tokens
$0
V4 Flash output per 1M tokens
OpenAI compatibility
DeepSeek's API supports both OpenAI and Anthropic formats. Switch from OpenAI by changing the base URL to https://api.deepseek.com and swapping your API key. No code rewrite needed. The model IDs are deepseek-v4-pro (flagship) and deepseek-v4-flash (economical).
| Notes | |
|---|---|
| AWS Bedrock | Managed US hosted infrastructure; ideal for enterprises already in the AWS ecosystem |
| OpenRouter | Unified API for 100+ models; useful for comparing models side by side |
| Together AI | US hosted infrastructure; good for data sovereignty concerns |
| Fireworks AI | Fast inference optimized for low latency applications |
| Groq | Ultra fast inference on custom LPU hardware |
| Ollama | Run locally on your own machine; supports distilled R1 variants |