Side by side pricing and context for DeepSeek Reasoner and o4-mini. Rates are curated catalog values with last verified dates on each model page.
DeepSeek Reasoner
DeepSeek reasoning endpoint when you want chain-of-thought style work at DeepSeek rates. Watch output tokens: reasoning traces can dominate the bill.
Input $0.550 / Output $2.19 per 1M
Context 128K · Approx
o4-mini
Smaller reasoning model when o3 is too expensive. Use for math, planning, and agent steps where you need chain-of-thought style work on a budget.
Input $1.10 / Output $4.40 per 1M
Context 200K · Exact (o200k_base)
Quick take
- DeepSeek Reasoner is cheaper on input in this catalog snapshot.
- DeepSeek Reasoner is cheaper on output in this catalog snapshot.
- Both last verified on their model pages (2026-08-02 / 2026-08-02).
Rate table
| Metric | DeepSeek Reasoner | o4-mini |
|---|---|---|
| Input / 1M | $0.550 | $1.10 |
| Output / 1M | $2.19 | $4.40 |
| Cached input / 1M | $0.140 | $0.275 |
| Context | 128K | 200K |
| Tokenizer | Approx | Exact (o200k_base) |