Add corrections, implementation notes, pricing changes, or usage caveats for other readers.
Feb 24, 2026
Knowledge cutoff
2025-01-01
Input modalities
Output modalities
Capabilities
128,000 tokens
Recent tweets and retweets from Inception
This Minecraft dog is an AI agent built with @_inception_ai's Mercury 2.
I talk. It understands, decides, and acts in real time.
Mere intelligence is ubiquitous. NPCs and mobs need reflexes.
This is the missing piece for AI in games.
Video
Building voice agents?
Get in touch with our team 👉 inceptionlabs.ai/enterprise#…
Link
Enterprise – Inception
We are leveraging diffusion technology to develop a new generation of LLMs. Our dLLMs are much faster and more efficient than traditional auto-regressive LLMs.…
Smart enough to reason, fast enough for a phone call. On real voice agent prompts, nothing beats Mercury 2 on both.
The world's first reasoning diffusion LLM: full reasoning pass in <300ms at 1000+ tok/s on standard NVIDIA GPUs.
💬"The reasoning quality we need without…
Mercury 2 is now available on Azure Foundry.
The world's fastest reasoning language model, built to make production AI feel instant.
Mercury 2 doesn't decode sequentially. It generates responses through parallel refinement, producing multiple tokens simultaneously and…
We're hiring at Inception.
Diffusion LLMs generate tokens in parallel, not one at a time, so much of the stack underneath has to be built from scratch. We're looking for people who want to build that.
Open roles:
➡️MTS, Kernels
➡️MTS, Security Engineering
➡️MTS, RL…
Discuss this model
Add corrections, implementation notes, pricing changes, or usage caveats for other readers.