Finance Lora Qwen3 4b is a LoRA adapter designed for use with the Qwen3-4B-4bit model. It has been fine-tuned on the gbharti/finance-alpaca financial instruction dataset using the MLX platform. The tool is intended to be used with MLX-compatible libraries and can be integrated into workflows via libraries, inference providers, notebooks, and local applications. Instructions are provided for downloading the model from the Hugging Face Hub and using it with MLX, specifically by loading the base model and specifying the adapter path. The model card notes a reduction in test perplexity from 30.636 (base) to 5.446 (tuned) across three seeds, indicating the effect of fine-tuning on the specified dataset. The tool was produced by slm-training. There is no information provided about licensing, pricing, or specific user roles beyond its focus on financial instruction data and compatibility with MLX. No details are given about integrations beyond those mentioned for MLX and related workflows. Downloads are not tracked for this model. Further details about its intended audience or broader use cases are not specified.
Latest indexed changes and source events
bha6kar/finance-lora-qwen3-4b discovered by the PulseGate indexer
Other apps tracked under the same category.
