What it does
- choice questions with up to 255 named options;
- yes/no questions;
- ordered score questions;
- readout temperature folded into the language-model norm;
- read-twice serving for short text-only requests.
Quickstart
Install a compatible PyTorch/CUDA environment and SGLang, then use the included front end:
bash serve_knowline.sh /path/to/KnowLine-9B-Gen1 0 8080
The front end exposes /v1/systemone:
curl -s http://127.0.0.1:8080/v1/systemone \ -H 'Content-Type: application/json' \ -d '{ "model": "m", "state": "The parcel arrived damaged.", "questions": { "refund": {"type": "noul", "instructions": "Should the customer receive a refund?"}, "route": {"type": "choice", "instructions": "Which team should handle this case?", "criteria": {"billing": "Payments and refunds", "support": "Technical troubleshooting"}} } }'
The included knowline_server.py, knowline_engine.py, and knowline_layout.py implement the scoring front end used for the evaluation below. The default front end reads short text-only requests twice and reads the option-label probabilities directly.
How Gen1 was made
- Base model: Qwen/Qwen3.5-9B.
- Training: one epoch of LoRA SFT, rank 32, alpha 64, language-model linear layers, bf16, merged into bf16 weights. Training examples include decision tasks written and labelled by LLM teachers.
- Maximum training sequence length: 12,288 tokens.
- The readout temperature
T=1.18, fitted on held-out data, is folded into model.language_model.norm.weight; see CALIBRATION.json.
The training data is not included in this repository. The checkpoint, tokenizer, chat template, inference front end, and checksums are included.
Evaluation
Full public Decision Index 0.3
Measured on 2026-10-25 with the released weights and corrected bf16 front end, read twice enabled:
Table with columns: Measurement, Result| Measurement | Result |
|---|
| Public Decision Index | 65.67 |
| Public raw index | 73.88 |
| Requests / scoreable requests | 140,620 / 140,178 |
| Scoreable answers / errors / unsupported | 140,178 / 0 / 0 |
| Sequential latency, median / mean | 96.7 / 207.7 ms |
The full suite was sent fresh to two H20 servers with 16 client shards. Two initial HTTP 504 transport failures were retried with unchanged payloads, weights and serving; no successful answers were retried based on correctness. Official scoring excludes 442 requests. No input was truncated. A separate sequential pass of 300 requests compared 477 answers with this model's own full run: 476 agreed, and the sole disagreement was a near-tie (margins 0.006 and 0.021). These are H20 measurements, not the board's hardware or private Full score.
Disclosures
- Decontamination: the training file was screened with standard and strict decontamination against the public DI suite and the public benchmarks we report. This procedure does not imply absence of pretraining exposure in Qwen3.5.
- Public suite: the Decision Index public suite is public; the board's private tests may show a lower score.
Calibration
CALIBRATION.json records the folded readout temperature. The front end uses temperature 1 after the temperature has been folded into the norm. Do not apply the same temperature a second time.
Limitations
- No image training was used in this release; image performance is zero-shot.
- The model is not trained with block-bidirectional attention.
- Long requests are subject to the model and serving context limits; the front end does not silently truncate a request.
- The model should not be treated as a general-purpose chat assistant. It is a structured decision readout model.
Files and license
The model weights and code are released under Apache 2.0. SHA256SUMS covers every file in this repository. The Qwen3.5-9B base model remains subject to its own license and terms.
Citation
@misc{knowline9bgen1, title = {KnowLine-9B-Gen1}, author = {PelaAI}, year = {2026}, howpublished = {Hugging Face model repository}}