AI you can count on.
We build AI to take tasks off your plate and help you be more free, expressive, and imaginative. Our aim is to support people and their work.
Overview
Deriva is our current development priority. Hikaru and PICO are also in development, at a slower pace. Every other series is a concept: development has not started. The status on each card shows where that series stands.
Our Main Lineup is the core family we plan to build around: OBIKI, Arcus, and Deriva. Being part of that lineup does not mean a model is already in development. Our Extras explore other uses and give us room for research and practice. Our Differentiators focus on capabilities such as image generation and speech.
No Sential model is publicly available yet. We are targeting for the first full public versions of Deriva and Kova. The date may move. We share development updates and selected demonstrations as the work progresses.
Our Main Lineup
OBIKI Series
OBIKI. The flagship
Our most ambitious series by far: a massive, fully custom model we intend to be genuinely frontier, for coding, bioscience, and cybersecurity.
OBIKI is the furthest-out concept in our lineup and the north star the rest of the work points toward. Built from the ground up, not finetuned.
Arcus Series
Arcus. The hardcore agent
A large model dedicated to code. The aim is depth: a system that is exceptional at software engineering rather than passable at everything.
Notes and architectural thinking exist, but no development has begun. We would rather build one thing well than many things averagely.
Deriva Series
Deriva. The everyday generalist
Our lead priority. A general-purpose model with vision, currently using Qwen3.8-27B as a candidate base while we evaluate the final choice. Capable across a wide range of tasks, including reasonable competence at code, but with no single specialty.
Deriva is the series we expect most people to use most of the time, the dependable everyday model in the lineup, and the model Kova is built around.
Our Extras
Hikaru Series
Hikaru. The testbed
Our first model. A finetune of Ministral-3 (8B), built to validate our ERL training method at a manageable scale. Still moving, just at a reduced pace while Deriva gets most of our attention.
Hikaru exists to answer a single question: does a model trained with explanatory reinforcement learning behave measurably differently than one trained with a scalar reward baseline?
Nero Series
Nero. For the physical world
A concept for robotics and wearables, intended to unify vision, voice, and language with the orchestration needed to act in the real world.
Nero is intended to focus on persona, tooling, and orchestration. The things a model needs when it lives in a body rather than a chat window.
Hirose Series
Hirose. The creative generalist
A powerful generalist built for creative work, with the range to help make real things across writing, art, and ideas.
There is a firm line with Hirose: it is meant to support human creativity, not replace it. If it ever starts taking work from human creators, we retire it, a rule that holds for every model we make.
Yuki Series
Yuki. Small and efficient
The smallest, leanest model we can build, made to punch far beyond its weight and stay startlingly capable for its size.
An efficiency project we build partly for the joy of it. Yuki (雪) means snow in Japanese: small, light, and everywhere.
Differentiators
Photon Series
Photon. Image generation
Our concept for image generation. Photon is intended to turn a written prompt into an image; Photon-1 is the planned first model.
Photon is our entry into visual generation, the counterpart to the reasoning models in the rest of the lineup.
PICO Series
PICO. Speech to speech
A speech-to-speech model: voice in, voice out, with no detour through text. In development, at a reduced pace, as PICO-1.
By skipping the text step, PICO aims for conversation that feels immediate rather than stitched together from separate parts.