Inception Labs

Diffusion language models for speed

Tags Developer tools API integration AI coding AI code assistant AI app builder

What Inception Labs is

Inception Labs is a company building diffusion-based large language models. Its core product is the Mercury series, which uses a parallel generation mechanism aimed at higher generation speed, with OpenAI-compatible APIs, on-premises deployment and fine-tuning support.

The company was founded by academics from Stanford, UCLA and Cornell, and states throughput figures and speed comparisons against mainstream models.

What you can do with it

Call Mercury models through an OpenAI-compatible API
Use them for code completion
Deploy on your own infrastructure
Fine-tune for specific tasks

Who it is for

Teams needing fast code generation
Enterprises wanting low-latency model APIs
Organisations requiring on-premises deployment

What to watch out for

Throughput figures and speed comparisons are vendor claims; benchmark on your own workload
Diffusion language models are a newer approach; evaluate quality, not only speed
On-premises deployment shifts infrastructure and security work to you
Confirm current model availability, pricing and licence terms