SOPHIA XT
Roadmap

Done, now
and next.

Seven stops. Four are finished, two are being built, and one is waiting on funding. We do not print dates we cannot keep, so the order is the plan.

Done

Public, and checkable today.

Open weights

Cassandra T1

1.33B masked-diffusion model. Weights, checkpoint and SHA-256 are public under Apache 2.0.

Open →
Measured

XE5 and anchor masking

147.5M parameters, 35 specialists. Anchor-token masking beats random masking by 1.67× held out, with a DOI and 504 published generations.

Open →
Live

Mnema V1

A 360M compositional edge model with sealed specialists, running in the browser.

Open →
Public

Sophia Stack X

A website platform for language models to run: one Rust binary, 53 MCP tools, signed releases. Live with a demo.

Open →
Now

Being built on our own hardware.

In build

Sophia S0

430M masked-diffusion MoE: one shared expert and 16 routed experts, top-2. The first step to Sophia-4B.

Open →
In build

SRW Pocket

Sophia Reader-Writer, 431M parameters: a 16-layer MoE reader and a dense writer. 291 MB on a phone.

Open →
Next

Planned, with a budget and written pass criteria.

Planned

Sophia-4B

3.88B total, 1.24B active, 256K context. Open weights and a hosted API. About $200K on rented H100s.

Open →
How we decide the order

Each stop has to
prove the next one.

S0 shows the architecture at a size we can train and test quickly. SRW Pocket tests it on a phone with no network. Sophia-4B is the same design at 3.88B. Pass criteria are written before each run and the result is published whichever way it goes.