Reka Flash 3
by NanoGPT · family reka · listed Mar 2025RW
Reka's 21B reasoning model for chat and coding, served directly by Reka.
Current rates USD per 1M tokens
Input$0.105
Output$0.21
Cache read—
Cache write—
Against the market
Input price vs. maker flagships ($/1M tokens)Output price vs. maker flagships ($/1M tokens)On the record
- Model id
reka/reka-flash-3- Context window
- 66K tokens
- Max output
- 66K tokens
- Input modalities
- text
- Output modalities
- text
- Reasoning
- yes
- Tool calls
- no
- Open weights
- yes
- Knowledge cutoff
- Jan 2025
- Released
- 2025-03-10
- Last updated
- 2025-03-10
← All NanoGPT listings