HawkTalk · Research
technical note · measured on-device

Grow, don't download.

Reconstructing a working on-device model from a sub-4 MB seed — and verifying it, chunk by chunk, against a signed root before you trust it.

Feedseed / Gemma3-270M 31-case FC eval CPU, greedy, byte-exact base 2026-07-26
A fine-tuned model is a small delta on top of a base most of its cousins already share. If the base is amortized across a fleet, you don't have to ship the whole model — only the delta, and only as much of it as quality actually needs. We measure that delta for Feedseed, HawkTalk's 270M on-device function-caller. Result: full function-calling quality from a 3.85 MB seed (50× smaller than shipping the delta, 65× smaller than the checkpoint), or 95% from 707 KB. We then close the honest gap — a reconstructed model must be proven correct — with a signed hash-linked Merkle log that both dates the version history and pinpoints exactly which reconstructed chunks came out wrong, so a node re-pulls only those and verifies each against a publisher signature before patching.

01 The idea

Downloading a 262 MB model to every device is the naive plan. But Feedseed isn't 262 MB of new information — it's a base scaffold (a vocab-trimmed Gemma3-270M that already knows English and how to pick a tool) plus a comparatively tiny, task-specific refinement. Ship the scaffold once, amortized across every app and variant that shares it; ship each model as its DNA — the delta — and let the device grow the full model back on the NPU.

The whole delta lives in 126 attention/MLP matrices (100.27M params). The 30.78M embedding + norm params have exactly zero delta — they're 100% in the shared base and never travel in the DNA. That's the first free win: a third of the model is, by construction, not part of the payload.

3.85 MB
DNA for 100% of full FC quality (r32, nf4)
707 KB
DNA for 95% quality (r2, int8)
50×
smaller than the 191 MB full delta
0
embedding bytes shipped — all in the base

02 Method

03 The frontier measured

Every configuration is scored against the full merged model under one greedy CPU harness (tool-selection 26/31, strict tool+slot 22/31, valid JSON 31/31). The base scaffold alone — zero DNA — already matches the full model on which tool (26/31); the DNA's entire job is argument/slot refinement, which moves the strict score from 16/31 back to 22/31.

configDNA payloadtool-selstrict (tool+slot)
FULL merged (reference)191.25 MB26/3122/31
base only — 0 DNA026/3116/31
r2-int8  — efficiency knee707 KB28/3121/31
r4-nf4492 KB24/3118/31
r16-int42.05 MB24/3119/31
r32-nf4  — full-quality knee3.85 MB26/3122/31
r64-int814.74 MB26/3122/31

SVD energy captured by rank: r1 29% · r2 41% · r8 69% · r16 82% · r32 95% · r64 99%. Full argument fidelity tracks ~95% energy → rank ~32.

⚠ Read the frontier honestly — it is not a smooth curve.
Rank-1 hurts. Rank-2 is an anomalously good projection (95%). Ranks 4–16 sag back. Full quality only returns at rank ~32. SVD orders directions by weight-energy, not task-relevance, so partial reconstructions bounce around. And on 31 cases, the difference between 18, 21, and 22 is 1–4 cases — inside eval noise. Treat the shape, not any single cell, as the signal. The sub-MB knee is real but rides one lucky low-rank direction; it is not a broad guarantee.

04 The verify-and-correct layer 26/26 tests pass

Inflation is only half an install. A grown model must be proven to be the real one — otherwise a bad node hands the next hiker a poisoned build with your name on it. We use one primitive for three jobs: a signed, hash-linked Merkle log (Git + Certificate Transparency, not a blockchain — no consensus, no coin).

✓ Verified end-to-end
In test, a rebuild corrupted at chunks [1,3,4] was identified as exactly [1,3,4]; forged correction bytes were rejected by their Merkle proof; genuine corrections patched and the result hashed to the signed root. Tamper-a-block, wrong-publisher, and wrong-key all rejected. 26/26.

05 Honest limits

06 Why it matters

Chapter one was on-device AI from a URL. This is chapter two: grow the model from a seed. A hiker on a recycled phone with no signal gets a trusted trail model by pulling kilobytes from the person next to them — and the same hashes that make that seed tiny are the ones that prove it wasn't tampered with on the way. Small enough to spread, signed enough to trust.