Haku Lab

Tag

ternary

← all experiments · all tags

2026Ternary · slice adapters · LoRA · ANE

Slice adapters vs LoRA on a ternary model

Retrain one slice of a ternary model's feed-forward channels in place, instead of adding a LoRA side path. Compared on four ternary models, timed on the M1 Neural Engine.

Read →
20267 failures · 6 models · no gradients, no training

Fixing small models by editing their weights: 5 of 7 failures

An MIT team fixed a looping model by editing its weights, with no training. We tried the same idea on 7 failures in 6 of our small models, with a random edit of the same size as the control.

Read →
2026Ternary 27B · agent battery · 21 traps

Does the 27B Bonsai actually deliver 27B?

Prism shipped a 27B that fits a laptop. We put it through a battery of agentic traps to see if it reasons like its size.

Read →
2026Ternary 27B · 7.17 GB · 40 tok/s on A10

Running a ternary 27B on a single A10

A 27B model whose weights are all −1, 0 or +1 — what it takes to run, and how fast it goes.

Read →
20269M ternary encoder · ANE · on-device

A 9M ternary encoder for accessibility-tree elements

One layer of the filter between raw accessibility data and the agent — element routing, on the Neural Engine.

Read →
202627M · structured · tool-calls

Structured tool-calls from a 27M model

How small can a model be and still learn a structured capability?

Read →
2026Ternary · on-device · field log

A few months of ternary experiments

Where ternary {−1, 0, +1} weights are the right tool for small, on-device models — and where they aren't.

Read →