This repository is not a GGUF or Transformers checkpoint. It is a Hologram key object: the weights of allenai/OLMoE-1B-7B-0924-Instruct re-encoded into Hologram's content-addressed .holo format so they stream, one verified block at a time, into Q, the on-device brain of the Hologram web OS. It runs in the browser on WebGPU, serverless, with nothing to install.

How it streams

The object is laid out for cold streaming from an untrusted CDN:

File	Role
`manifest.json`	the root. Names every tensor and the key (content hash) of its block.
`b/sha256_*.gz`	the tensor blocks. Each filename is the SHA-256 of its bytes.
`tokenizer.gguf`	bundled header (where present), so loading is fully serverless.

Q fetches the manifest, then pulls each block by its key and re-derives sha256(block) on arrival. If a byte is wrong, the block is rejected. Nothing is trusted; everything is proven.

Verify (Law L5)

The object's identity is the SHA-256 of its manifest, pinned in Q's catalog before a single byte of weight is trusted:

did:holo:sha256:9cf97ec1c761fd4ef51bc0cd4ac37a0cd8eaa11f1b19b3ae6a141486ad3fe5ad

bash

# the manifest hash equals the pinned identity above
curl -sL https://huggingface.co/HOLOGRAMTECH/q-olmoe-1b-7b/resolve/main/manifest.json   | sha256sum

Specifications


Architecture	OLMoE (64 experts, 8 active)
Precision	q4 · MoE
Object size	3.6 GB
Hidden size	2048
Layers	16
Heads (Q / KV)	16 / 16
FFN	1024 x 64 experts
Vocab	50304
Context	3000
Format	`holo-2bit/1`

Provenance and license

Derived from allenai/OLMoE-1B-7B-0924-Instruct. The re-encoding is lossless-by-construction at the key level: every block is content-addressed, so the object either re-derives to its pinned identity or it is refused.

Run it

These weights load through Q, not a standard runtime. Open the Live Space or visit gethologram.ai to run Hologram, then pick OLMoE-1B-7B from Q's model list.

_{Composed on the golden ratio. One key, everything.}