Inception
About Inception
Inception develops language models using diffusion-based generation. Its Mercury model family is presented as a language-model line that can produce groups of tokens through a different generation process from conventional autoregressive decoding. Provider performance claims require their own declared test conditions.
Focus
- Diffusion language models
- Model generation research
- Hosted model interfaces
Projects and interfaces
Mercury
diffusion model family
Inception's published language-model family based on diffusion.
Scope and limitations
- Architecture descriptions and provider-reported speed claims are separate from measured Sigma5C results.
Official references
- Inception official website. Checked .
Models
3 named catalog records. These are name and distribution records, not a count of distinct model architectures.
- inception/mercury-2
- inception/mercury-2.5
- inception/mercury-3