Side by side pricing and context for DeepSeek R1 and DeepSeek R1 (Fireworks). Rates are curated catalog values with last verified dates on each model page.
DeepSeek R1
Full DeepSeek R1 reasoning model for hard STEM and planning tasks at DeepSeek rates. Watch output tokens; reasoning traces can dominate spend.
Input $0.550 / Output $2.19 per 1M
Context 128K · Approx
DeepSeek R1 (Fireworks)
DeepSeek R1 on Fireworks for low-latency reasoning-class chat. Validate Approx counts and compare native DeepSeek R1 before locking spend.
Input $0.900 / Output $0.900 per 1M
Context 128K · Approx
Quick take
- DeepSeek R1 is cheaper on input in this catalog snapshot.
- DeepSeek R1 (Fireworks) 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 R1 | DeepSeek R1 (Fireworks) |
|---|---|---|
| Input / 1M | $0.550 | $0.900 |
| Output / 1M | $2.19 | $0.900 |
| Cached input / 1M | $0.140 | n/a |
| Context | 128K | 128K |
| Tokenizer | Approx | Approx |