Hermes Parallelization v1
A Laya typed-decision model, fine-tuned with System One Studio on an Apple silicon Mac. It answers the questions below in a single forward pass, with calibrated probabilities and zero generated tokens.
Base model: convaiinnovations/laya · method: lora, proper objective · trained in
49 minutes on an Apple silicon Mac.
Measured on the held-out test split
| Metric | Base model | This model |
|---|---|---|
| Accuracy | 23.3% | 53.3% [36.1%–69.8%] |
| Calibration error (ECE) | 0.385 | 0.174 |
| Log loss | 1.421 | 0.982 |
| Brier score | 0.697 | 0.427 |
| Decisions scored | 30 | 30 |
Fine-tuning fixed 12 test decisions the base model got wrong and broke 3 it got right (exact McNemar p = 0.035).
Test rows were never trained on. Accuracy intervals are Wilson intervals; the paired test is an exact McNemar test between the base and the fine-tuned model on the same rows.
Use it
pip install systemonemodels
systemone pull vorthalis/hermes-parallelization-v1
import json
import laya_mlx as laya # pip install laya-mlx, on Apple silicon
from systemone import snapshot_download
path = snapshot_download("vorthalis/hermes-parallelization-v1")
agent = laya.load(str(path))
questions = json.loads((path / "questions.json").read_text())
print(agent.predict("your text here", questions)["answers"])
Ask it these questions: the instructions and option texts are part of the model's input, so changing them changes the task it was tuned for.
{
"parallelizable": {
"type": "noul",
"instructions": "Can run in parallel with other tasks?"
},
"dependency_risk": {
"type": "score",
"instructions": "Rate dependency risk",
"criteria": [
"none",
"low",
"medium",
"high"
]
}
}
The same folder also loads in the upstream PyTorch laya package on Linux and NVIDIA, and
System One Studio can export it to ONNX.
NoulXP: this version carries a NoulXP package in noulxp/ (noulxp/0.1, encoder-markers profile: an ONNX graph over these weights). Before publishing, System One Studio recorded this fine-tune's own answers to NoulXP's 52 conformance requests with the laya package 0.3.5 (float32, CPU), and the NoulXP reference runtime (noulxp 0.4.0) reproduced 52 of 52 cases on the CPU (max |Δp| 5.0e-05, tolerance 0.01). systemonemodels.tech runs the same check before it shows the model as NoulXP compatible.
Provenance
{
"base_model": "convaiinnovations/laya",
"hyperparameters": {
"method": "lora",
"objective": "proper",
"epochs": 4,
"batch_size": 8,
"grad_accum": 2,
"lr": 0.0002,
"head_lr": 0.0001,
"lora_rank": 16,
"lora_alpha": 32,
"lora_dropout": 0.05,
"dora": false,
"rslora": false,
"loraplus_ratio": 1.0,
"lora_layers": 0,
"full_layers": 4,
"head_dropout": 0.1,
"weight_decay": 0.01,
"warmup": 0.06,
"max_grad_norm": 1.0,
"shuffle_options": true,
"class_weighting": "none",
"patience": 2,
"grad_checkpoint": "auto",
"precision": "bfloat16",
"seed": 13
},
"train_decisions": 242,
"best_epoch": 4,
"temperature": [
1.6369030475616455,
5.0,
0.8773
],
"temperature_by_options": {
"choice:3-5": 1.7601518630981445,
"choice:6-10": 1.0000158548355103,
"choice:11+": 0.10058280825614929,
"choice:2": 1.9063563346862793
},
"dataset_sha256": "9de0bc947c1d6a64c7187f62fa1cf2cf3b4abddc5f5562c56d17ae9df14d3c59",
"trained_on": "2026-10-14T08:03:13"
}
License and attribution
Apache-2.0. Laya and its pretrained weights are by
Convai Innovations; this checkpoint is a
fine-tune of convaiinnovations/laya and carries the same licence. Fine-tuned and published with
System One Studio.