4.2–3B SMALL MODEL. LOCAL COMPUTE.

Big ideas.
Your little GPU.

Meet Nanbeige. A language model that runs right here, on your device. Ask it something. See how fast it thinks.

Your prompts stay here. No account. No server inference.

THE MODEL

Nanbeige4.2–3B

ON DEMAND

Model profile

EngineContext

Compatibility modeNo subgroups · compatibility profile

First download: about 2.35 GB. Cached when browser storage allows.

Give it something to think about.01 / PROMPT

Try

Answer limit

Ready when you are

02 / RESPONSEALL YOURS

A little patience.
Your answer will appear here.

THE SPEEDOMETER LOCAL

How fast is yours?

Every answer is a benchmark.
These numbers come from your device.

—

tokens / secondGeneration speed

—Time to first token

—Prompt tokens

—Output tokens

—Total run time

Run your first prompt to get a reading. Loading time is measured separately.

Session history

No runs yet. Your GPU is waiting.

JSON includes prompts, output, per-token timings, and device details. Saved only when you export.

Take a closer lookRepeatable benchmark & validation suite

One stack. Two passes.

22 shared transformer layers run twice, with separate attention history for each pass. The trained two-pass configuration stays intact.

22 LAYERS↻2 PASSES

Q4 weights, FP32 activations, and 44 separate KV-cache slots in the custom engine.

WebGPU is required. Phones and tablets have not been verified on physical devices and have tighter memory limits. Start with 512 context; larger contexts use more GPU memory. Results vary with browser, temperature, and other running apps.

Big ideas.Your little GPU.

Nanbeige4.2–3B

Big ideas.
Your little GPU.