Context
252K
Input
$0.20/M
Output
$0.75/M
Aliases:mercury-2-5mercury2.5inception-mercury-2.5
Mercury 2.5 is Inception's fast reasoning diffusion LLM (dLLM). Instead of generating tokens one at a time, it produces and refines many tokens in parallel, for low-latency chat, agents, and structured workflows.
InceptionDeprecating on Inception
inception/mercury-2.5QuantUnspecified
- Input
- $0.20/M
- Output
- $0.75/M
- Context
- 252K
- Max out
- 66K
- Cache read
- $0.02/M
- Cache write
- $0.20/M
- Thinking out
- $0.75/M