---
license: unknown
pipeline_tag: text-classification
base_model:
- Qwen/Qwen3-4B-Instruct-2507
- Qwen/Qwen3.5-2B
tags:
- opensysone
- decision-scoring
- lora
- experimental
---

# OpenSysOne

**Inspired by [Jev](https://typesafe.ai/), TypeSafe.ai's System One model.**
Credit goes to the TypeSafe team for inspiring this project's exploration of
structured decisions with probabilities. OpenSysOne is an independent experimental implementation; API
compatibility does not establish Jev equivalence.

The **completed 4B release** scores a state, question and explicit candidate
answers, returning probabilities over those choices. Training, separate
calibration, final evaluation and local API verification completed on
17 September 2026.

Start with the [model and reconstruction notes](model/README.md),
[results report](results/report.md), or [publication guide](docs/README.md).
The calibrated artifact is [model/model.pt](model/model.pt).
It contains custom OpenSysOne adapter/head weights and metadata. The pinned
Qwen3-4B-Instruct-2507 base is required separately; this is not a standalone
Transformers model or a standard PEFT adapter package.

## Measured results

The full comparison uses the unchanged pretrained yes/no verifier, with a separate
temperature fitted for each model. Intervals are paired 95% source-group bootstrap
intervals for selected minus base accuracy.

| Evaluation | Decisions | Selected | Base verifier | Accuracy gain (95% interval) |
| --- | ---: | ---: | ---: | ---: |
| Known-family test | 2,042 | 92.90% | 84.48% | +8.42 pp [6.85, 9.89] |
| Social IQA family holdout | 768 | 72.92% | 70.31% | +2.60 pp [0.13, 5.34] |

On a separate matched 320-decision profile, selected accuracy was 89.06%, versus
80.94% for the base verifier and 86.25% for a base model using one constrained
answer-label token. The selected scorer was **slower on all 12 profiled workloads**:
1.11–1.17× the verifier latency and 2.18–15.58× the label baseline latency.
These are warm, serial FP32 measurements on one GB10, not concurrent-serving
throughput or comparisons with generated reasoning. See [tables and charts](results/).

## Model and limits

The release uses rank-8 additive adapters and a scalar head: 16,517,633 trainable
parameters. The selected expanded branch's step 0 retains the refinement parent's
step-1,500 weights. The expanded branch's later step 159 was not selected; its
post-selection diagnostic gains are reported separately.

One temperature, 1.745822, was fitted on 510 separate known-family calibration
examples. Social IQA calibration remains limited: its top-label ECE is 8.30%.
No general intelligence, Jev-level quality or universal calibration claim follows.
Four-choice Banking77 is not the full 77-label task; benchmark grouping does not
rule out base-model pretraining overlap. The 1,024-token inference limit includes
the complete formatted candidate prompt, and longer inputs are rejected.

## Files and provenance

- [model/](model/README.md): calibrated artifact, hash and pinned base requirements.
- [docs/](docs/README.md): layout and [reproduction guide](docs/reproduce.md).
- [source/](source/): complete committed project source, tests and usage guides.
- [results/](results/): final metrics, profiling report, tables and charts.
- [archive/](archive/README.md): index to preserved experiment history.

The original pointers remain authoritative:
[FINAL_MODEL.json](FINAL_MODEL.json) identifies the calibrated release,
[PROFILE_RESULTS.json](PROFILE_RESULTS.json) identifies verified profiling and
wrap-up evidence, and [CURRENT_SNAPSHOT.json](CURRENT_SNAPSHOT.json) identifies
the earlier training backup. Historical payload paths and hashes are preserved.

The existing `license: unknown` metadata is unchanged. Base-model and dataset
licenses remain separate; see the source's
[original data provenance](source/results/public-decisions-v1-manifest.json) and
[expanded data provenance](source/results/20260917-expanded-data/dataset-manifest.json).
Base weights and credentials are excluded. Hosted Jev calls require separate
authentication and were not exercised in this evaluation.