DeepSeek vs Claude API pricing
Checked against the official price pages on 2026-10-12
In short
At 100M input and 20M output tokens a month, the cheapest model from each is deepseek-flash (about ¥270 a month) and Claude Haiku 5.5 (about ¥135 a month). Claude's entry model is cheaper; DeepSeek's costs about 2.0 times as much.
The most expensive from each is deepseek-v4-pro (about ¥1,080 a month) and Claude Fable 5.1 (about ¥13,466 a month). Price says nothing about whether the models are equally capable; weigh it with benchmarks and context length.
For your own usage: API cost calculator
All models
Prices are per million tokens: ¥ is CNY, $ is USD; time-of-day pricing is written peak / off-peak. Monthly cost is in CNY and assumes 100M input and 20M output tokens a month, no cache hits and half of calls at peak hours where that applies, cheapest first. Models priced in USD are converted at 1 USD = 6.733 CNY (PBoC central parity, 2026-10-09).
| Model | Input | Cache hit | Output | Monthly (CNY) |
|---|---|---|---|---|
| AnthropicClaude Haiku 5.5 | $0.1 | $0.01 | $0.5 | 135 |
| DeepSeekdeepseek-flash | ¥2 / 1 | ¥0.04 / 0.02 | ¥8 / 4 | 270 |
| DeepSeekdeepseek-v4-pro | ¥9 / 4.5 | ¥0.3 / 0.15 | ¥27 / 13.5 | 1,080 |
| AnthropicClaude Sonnet 5.5 | $2 | $0.1 | $10 | 2,693 |
| AnthropicClaude Opus 5.5 | $4 | $0.2 | $20 | 5,386 |
| AnthropicClaude Fable 5.1 | $10 | $0.25 | $50 | 13,466 |
Price changes at both vendors
DeepSeek V4-Pro launched at a list price of ¥12 input (cache miss) / ¥24 output, with a 75% launch discount (¥3 / ¥6). source
DeepSeek cut cache-hit input prices across the lineup to a tenth of launch levels: ¥0.02 for V4-Flash, ¥0.025 for V4-Pro. source
DeepSeek said V4-Pro would stay permanently at a quarter of its list price after the promotion — ¥3 input / ¥6 output (about $0.435 / $0.87). source
DeepSeek introduced peak/off-peak API pricing: weekday 9:00–12:00 and 14:00–18:00 Beijing time are peak; off-peak is half price. V4 Pro output ¥27 peak / ¥13.5 off-peak; V4 Flash output ¥9 / ¥4.5. source · our coverage
DeepSeek launched V4.1 Flash and said it planned to route V4 Pro requests to the new model at Flash prices. source · our coverage
Other comparisons
- DeepSeek vs Kimi
- DeepSeek vs Qwen
- DeepSeek vs Zhipu
- DeepSeek vs MiniMax
- DeepSeek vs OpenAI
- DeepSeek vs Gemini
- Kimi vs Qwen
- Kimi vs Zhipu
- Kimi vs MiniMax
- Kimi vs OpenAI
- Kimi vs Claude
- Kimi vs Gemini
- Qwen vs Zhipu
- Qwen vs MiniMax
- Qwen vs OpenAI
- Qwen vs Claude
- Qwen vs Gemini
- Zhipu vs MiniMax
- Zhipu vs OpenAI
- Zhipu vs Claude
- Zhipu vs Gemini
- MiniMax vs OpenAI
- MiniMax vs Claude
- MiniMax vs Gemini
- OpenAI vs Claude
- OpenAI vs Gemini
- Claude vs Gemini