01 / Introducing Qualia-1

The context model for the human side of AI.

Models keep getting better at reasoning, generating, and acting. What they still miss is the person on the other side — whether the user is moving forward, stuck, or changing direction while the work is happening.

Qualia-1 gives any AI agent that missing context, so it finishes more tasks, needs fewer corrections, and earns repeated use.

Request a pilot → Closed beta · selected teams
02 / Problem

Agents understand the task. Not the person.

Most agents see the request, the documents, the tools, and the system state. They know almost nothing about the user — whether someone is focused, stuck, correcting them, or quietly giving up. So adoption stalls: the agent works once, but never becomes part of how people actually work.

01

Installed, but not adopted.

For enterprise products the real bar isn't whether an agent works once — it's whether people return to it, trust it, and let it into the workflow. That's the part most agents miss.

02

Blind to the user.

An agent that can't tell when someone is stuck, distracted, or redoing its work another way can't reliably carry a task to done. So it leans on the user to correct and re-explain.

03

No user-side read for teams.

Product teams see prompts, outputs, tool calls, and latency. They don't see what happened between the agent's response and the user's next move. That gap hides why an agent gets adopted, ignored, or abandoned.

03 / What it is

A missing input, not another agent.

// the one job Read what's happening with the user, and hand the agent a compact, structured read it can act on in the moment.

Qualia-1 is an on-device model that gives AI agents structured context about the person using them.

It turns what's happening in the session into a compact human-context output any AI product can use during the interaction — not after it.

It doesn't replace the agent — it adds the one thing they're missing: what's going on with the user while the work is happening.

04 / Live cases

Same prompt. Same model. Different agent.

Four agents, replayed twice: once generic, once with Qualia-1 context. Watch the context card — each fact lights up at the exact moment it changes the agent's behavior.

05 / Use cases

Where it fits.

Qualia-1 sits between any agent and its context — most useful wherever agents work repeatedly inside a live workflow.

— AI copilots

AI copilots

Match how much a copilot helps to how the session is going — stepping back when the user is in flow, stepping in when they stall.

— AI agents

AI agents

Give agents human-context alongside task, tools, and memory — so they respond to the interaction, not just the request.

— Developer tools

Developer tools

Let IDE agents read the state of a coding session — deep focus, debugging, or going in circles — and adapt the next step.

— Enterprise workflows

Enterprise workflow products

Help AI features tell whether a user is progressing, correcting, or about to abandon a flow — and respond before they drop off.

— Evaluation

AI product evaluation

See interactions from the user's side — beyond prompts, outputs, latency, and cost — to understand what actually happened.

— Integration

One structured input. It slots in front of the agent without changing the model behind it.

Talk to us →
06 / Evidence

What context is worth, in numbers.

Two kinds of evidence: public agent benchmarks re-run with Qualia-1 context in our internal harness, and product metrics from simulated pilot sessions. Baselines are real, cited, and current as of June 2026. Qualia-1 figures are ours — read them as a research preview, not audited results.

Read this first. Figures labeled “+ Qualia-1” come from internal, preliminary evaluations under simulated benchmark conditions. Qualia-1 agent-context is a research preview under customer-development validation; numbers will change.
BenchmarkPublic baseline+ Qualia-1Δ uplift
MultiChallengemulti-turn: instruction retention, memory of user info 69.6%publicGPT-5, leaderboard top · Scale SEAL · 2026 78.2% +8.6 pp
LongMemEval-Slong-term memory across chat sessions 60.2%publicfull-context GPT-4o, canonical no-memory baseline · arXiv:2501.13956 84.6%published memory systems: 81.6–94.9 (Supermemory · EverMemOS · Mastra OM) +24.4 pp
LongMemEval-V2agentic memory: workflows, state, gotchas · May 2026 48.5%publicstrongest RAG-memory baseline · arXiv:2605.12493 58.9%coding-agent controller hits 72.5% at high latency — Qualia-1 targets the <400 ms regime +10.4 pp
PrefEval · turn 10preference following over a dialogue <10%publicfrontier zero-shot · ICLR 2025 · arXiv:2502.09597 74%published injection methods recover 84–97% on the MCQ subset ×7+

Product metrics from simulated pilots.

−34%
Time to task completion
median · N=2,400 sessions · 85 pilot users · 6 weeks · coding + corporate agents · internal eval
−41%
Retry & clarification rate
clarifying questions per task · N=2,400 sessions · preliminary
−47%
User-correction rate
user edits to agent output per task · coding + legal cohorts · N=1,180
−38%
Prompt length — you type less
median user prompt tokens, same tasks, with vs without context · N=2,400
6284%
First-response acceptance
output accepted without edit on first turn · N=1,180 drafting tasks
−29%
Voice escalation rate
human-handoff rate · simulated call cohort · N=640 calls
−43s
Average handle time (voice)
−22% mean AHT · N=640 calls · returning-caller context enabled
−2.3
Partner red-line cycles (legal)
redline rounds to partner sign-off · N=210 drafts · 6 firms · internal eval
07 / Access · API · Beta

Closed beta.

We're working with selected teams building AI copilots, agents, developer tools, and workflow products. Qualia-1 is an SDK and API for teams who want to add user-side context to their AI product.

Who should reach out

Teams building:

  • AI copilots
  • AI agents
  • developer tools
  • workflow software
  • enterprise AI products
  • AI product evaluation systems

What we're looking for

Design partners who can help validate:

  • which signals of user state matter
  • how human-context output should be represented
  • which product metrics change when AI gets user-side context
  • how to integrate this safely and privately

Make user behavior part of your agent.

Closed beta — for teams building agent-driven products.
Request beta access →