DeepSeek V4 Pro vs V4.1 Flash
Both have a 1M-token context and up to 384K tokens of output, and both work in Codex and Claude Code. They differ in price, image input and how many requests can run at once.
| DeepSeek V4.1 Flash | DeepSeek V4 Pro | |
|---|---|---|
| Model name | deepseek-flash | deepseek-v4-pro |
| Served by | DeepSeek-V4.1-Flash-1017 | DeepSeek-V4-Pro-0813 |
| Context | 1M | 1M |
| Max output | 384K | 384K |
| Image input | Yes | No |
| Concurrent requests | 2,500 | 500 |
- Flash costs less. Input is $0.3 per 1M tokens at peak, against $1.32 for Pro.
- Only Flash reads images. Pro is text only.
- Flash allows 2,500 requests at once; Pro allows 500.
- DeepSeek's own Codex setup uses deepseek-flash by default, with deepseek-v4-pro as the alternative.
| Model | Input, cache hit | Input, cache miss | Output | |
|---|---|---|---|---|
| DeepSeek V4.1 Flash | DeepSeek, peak hours | $0.006 | $0.3 | $1.2 |
| DeepSeek, off-peak | $0.003 | $0.15 | $0.6 | |
| HaryAI | $0.003 | $0.135 | $0.54 | |
| DeepSeek V4 Pro | DeepSeek, peak hours | $0.044 | $1.32 | $3.96 |
| DeepSeek, off-peak | $0.022 | $0.66 | $1.98 | |
| HaryAI | $0.022 | $0.594 | $1.782 |
Questions
Can I switch between them?
Yes. It's one setting: the model name, deepseek-flash or deepseek-v4-pro.
Where did V4 Flash go?
It was retired. Its old names still work and are served by V4.1 Flash at the Flash price.
Sources
DeepSeek's figures are from its own docs, checked on 2026-11-15.
DeepSeek at 55% off the official price
Cheaper than DeepSeek's own off-peak price, around the clock.