Skip to content

Model / perceptron

Perceptron Mk1.5

Perceptron Mk1.5 is Perceptron's embodied reasoning model for physical agents. It accepts text, image, video, and audio input, and answers with text plus optional structured annotations: points, boxes, polygons, tracks,...

Know · from the live catalog

Model id
perceptron/perceptron-mk1.5
Maker
perceptron
Context window
37K tokens
Input price
$0.15 / 1M tokens
Output price
$1.50 / 1M tokens
Input types
text, image, video, audio
Output types
text
Tool calling
supported
Reasoning
supported
Listed
2026-09-25

Verify · per provider

Preview · not live

Provider bond

10,000 USDG required before live traffic

Quality canaries

Weights fingerprint, exact-answer set, empty-reply rate

Signed receipt

Ed25519 over model, provider, tokens, cost and hashes

TEE attestation

Required on the attested lane only

No provider is bonded yet. These checks show what each call to this model will carry once bonded providers go live.

Route · node
const answer = await client.chat.completions.create({
  model: "perceptron/perceptron-mk1.5",
  messages: [{ role: "user", content: "Hello" }],
});

answer.receipt; // signed by the router once receipts ship