spencernorton's picture
Release v1.6.0 (v9): genuine unassisted FunctionGemma-270M residual QLoRA adapter
71dce43 verified
|
Raw History Blame Contribute Delete
1.25 kB
---
license: gemma
base_model: google/functiongemma-270m-it
tags:
- mlx
- lora
- functiongemma
- tool-calling
- arrmes
---
# Arrmes FunctionGemma-270M v9 Residual FP16 QLoRA Adapter (`v1.6.0`)
Unassisted residual FP16 LoRA adapter over the 4-bit `FunctionGemma-270M` base (`183.1 MiB` total footprint), subject to the [Gemma Terms of Use](https://ai.google.dev/gemma/terms) (`GEMMA_TERMS_NOTICE.txt`).
![Arrmes v1.6.0 Accuracy Comparison](figures/arrmes-model-accuracy-comparison.png)
## Artifact & Unassisted Benchmark Summary (`LocalModel`, `backend="mlx-lm"`)
- **`v9/adapters.safetensors` SHA-256**: `a0337f083c362d66f406129d0ee6f4661597055bd51772233241c0e2690cf124`
- **Template-Disjoint `acceptance-v7` `v2` Positives (`N=150`)**: Track A Raw `126/150` (`84.0%`), Track B Verified `135/150` (`90.0%`, Wilson 95% LB `84.16%`)
- **Template-Disjoint `acceptance-v7` Hard Negatives (`N=150`)**: Track A Raw `49/150` (`32.67%`), Track B Verified `133/150` (`88.67%`, Wilson 95% LB `82.60%`)
- **Frozen `overnight-v1` (`N=80`)**: Track A Raw `59/80` (`73.75%`), Track B Verified `69/80` (`86.25%`)
- **Apple M4 Pro Metal Warm Latency**: `p50 = 110.08 ms`, `p95 = 124.80 ms` (`183.1 MiB` combined 4-bit base + FP16 LoRA footprint).