299B Helix × GPT-2 ▶ RECORDED RESULTS — real gate outputs, not live

The model you know, on a stack you can verify from the first byte.

This demo runs real, unchanged GPT-2 on Helix — a software stack that rebuilds itself from 299 hand-typed bytes, so every layer of it can be audited instead of trusted. The output was checked, token by token, against an independent referee. Pick your door:

THREE DOORS

Choose your depth

HONESTY UP FRONT

What this demo is — and isn't

IT IS

· The real, unchanged public GPT-2 (and SmolLM2), running on kernels compiled from Helix source by a compiler that rebuilds from 299 bytes powered by Helix

· Verified: every model matched an independent oracle token-for-token (25/25), gated fail-closed

· Honest about its sources: the badge always says LIVE, REPLAY (a real recorded run) or PREVIEW (mock)

IT ISN'T

· An assistant. GPT-2 is a 2019 base completion model — it continues text, sometimes repetitively. That's the real model, not a bug.

· Fast. Live GPT-2-XL runs at ≈10 s/token by design — the pitch is trust, not speed (it will get faster as Helix develops).

· Verified below PTX. One closed NVIDIA step is trusted-once — disclosed, never hidden.

DON'T TAKE OUR WORD

Re-check the core yourself — one command, one minute

# clean checkout · CPU-only · no GPU, no weights · fail-closed git clone https://github.com/Questeria/helix && cd helix bash scripts/reproduce_trust.sh # rebuilds the entire ladder from the 299-byte root and asserts: seed 9837db12 · fixpoint 0992dddd · gcc-DDC K1 84363adb

The model legs additionally need the public HuggingFace weights and an independent oracle — the repo's runbook walks through both tiers honestly.

Honest residuals: fp32 · verified to PTX, not SASS · single GPU (sm_86) · base models, not assistants · the oracle shares the model's spec. Every number on this page is a recorded, committed result. start here · guided run · expert · proof · models