ebnezr-isaac/qwen3-0.6b-traffic-signal-ft1
text-generation model by ebnezr-isaac
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
Qwen3-0.6B, QLoRA fine-tuned (r=16, 2 epochs) on 14,493 delay-aware junction decisions, trained in about 3.9 hours on an RTX 2060 (6 GB) and served int4 on the same GPU through Foundry Local. This is the student behind the dissertation's headline result. Result (simulation). On a London street grid (Bloomsbury, from OpenStreetMap) with demand set so the network can clear, this model cut mean vehicle delay by 18.3% against stock Qwen3-0.6B compiled through the identical int4 pipeline (ft0), winning on all 30 seeds (one demand pattern, varied driver behaviour). It matched the MaxPressure baseline to within 0.03% and beat a fixed-time plan by 8.8%. On the Euston Road corridor it did not help. Offline: 97.4% agreement with the teacher labels on the 1,510-state delay-aware holdout (stock: 55.1%), 100% valid JSON. lora/ holds the LoRA adapter (base Qwen/Qwen3-0.6B). Then drive a SUMO simulation with it from the project repo: The controller asks…
Read the full model card ↗ · Preview checked 2026-10-06T07:41:57.322Z
Inside the original model card — Document outline
- Qwen3-0.6B fine-tuned for traffic-signal phase selection (ft1, headline model)
- Use it with Foundry Local
- Limitations
- Training data
- Citation
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the model card
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
ebnezr-isaac/qwen3-0.6b-traffic-signal-ft1 is a text-generation model repository published by ebnezr-isaac on Hugging Face. The source reports the onnxruntime-genai library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-generation
- Library
- onnxruntime-genai
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- apache-2.0
- Access
- Not gated by Hugging Face
- Created
- 2026-10-06T06:08:03.000Z
- Last modified
- 2026-10-06T06:18:00.000Z
Compatibility and lineage
Language coverage was not reported.
Reported base models: Qwen/Qwen3-0.6B
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-06T06:39:56.505Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.