ModelMercury 2.5 LLM
Mercury 2.5
A diffusion-based language model developed by Inception Labs that employs a parallel denoising architecture to achieve non-autoregressive text generation, with high inference throughput as its defining characteristic.
Timeline (last 90 days)
Sep 23
The original Hacker News post about Mercury 2.5 did not provide details on model scale, evaluation environment, hardware configuration, or quality benchmarks.
Expired50%
Sep 18
In the Browser Use combined architecture, Jev is responsible for structured decision-making, another small language model (Mercury 2.5 in the demo) is responsible for generating text input, and the browser code is responsible for executing actions.
Unverified50%