DeepSeek free tier: quotas, limits and gotchas
DeepSeek Free Tier
Last verified: 2026-09-25
Models and quotas
| Model | Quota | Window | Notes |
|---|---|---|---|
| DeepSeek V4 Flash | The web version is basically free, but it is rate-limited during peak hours. | Rate Limiting | API requires additional payment |
| DeepSeek V4 Pro | The web version is basically free, but it is rate-limited during peak hours. | Rate Limiting | API requires additional payment |
What you get for free
| Feature | Free tier | Notes |
|---|---|---|
| Dialogue and programming | Available (with peak traffic limiting) | |
| Context length | Up to 1M tokens (API specification) | |
| Maximum output | 384K tokens (API specification) |
Limits at a glance
- Quota reset: Peak hours have real-time rate limiting, with no clear daily reset
- Commercial use: Allowed
- Conditions: The web version can be used without registration, API requires payment, available for users in Taiwan, no regional restrictions, and supports Chinese and English interfaces
Watch out for this
The free web version and API pricing are two separate systems. The API cost is very low but requires development skills. The web version's quota policy lacks official detailed explanation.
They also argued about these things on another page.When the free quota is insufficient, how much does it cost to pay for it?
| Plan | pricing | Description |
|---|---|---|
| API - DeepSeek-V4-Flash | API pay-as-you-go (usage-based) | Model name deepseek-flash (DeepSeek-V4.1-Flash, launched 2026-09-10; the legacy names deepseek-v4-flash / deepseek-v4-flash-vision-exp still work, but are served by V4.1-Flash and billed at the Flash price). Per million tokens (off-peak/peak): cache-hit input US$0.003/0.006, cache-miss input US$0.15/0.30, output US$0.60/1.20. Peak hours are 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday (Chinese public holidays excluded); peak prices are 2x off-peak. ⚠️ Price cut in September 2026: on 9/4 we recorded cache-miss input US$0.22 and output US$0.66 (off-peak). Read directly from the api-docs.deepseek.com English and Chinese pricing pages on 2026-09-25. |
| API - DeepSeek-V4-Pro | API pay-as-you-go (usage-based) | Model version DeepSeek-V4-Pro-0813. Per million tokens (off-peak/peak): cache-hit input US$0.022/0.044, cache-miss input US$0.66/1.32, output US$1.98/3.96 (read directly from the official site on 2026-09-25; prices unchanged). Concurrency limit 500 (Flash: 2500). |
Checked 2026-09-25
Information for Taiwanese users
- Payment method:Available, users should assess data export concerns on their own.
- Chinese support: full
- Available platforms:web
Related
Sources: DeepSeek official website · DeepSeek API Pricing Document
Affiliate Links Notice
