Downloads · 30 days
26
9% of all-time downloads
ibrahimhka/marketanalyst-ReAct-distill-Qwen3-8B-LoRA-GGUF
marketanalyst-ReAct-distill-Qwen3-8B-LoRA-GGUF is a machine learning model from ibrahimhka. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as mit.
This adapter was converted to GGUF format from a QLoRA fine-tune of Qwen3-8B, using llama.cpp's convertloratogguf.py.
Downloads · 30 days
26
9% of all-time downloads
All-time downloads
281
Public
Repo size
535 MB
Likes
0
Public
Click a slice to open those files.
.safetensors349 MB · 65%
From the Hugging Face model README
This adapter was converted to GGUF format from a QLoRA fine-tune of
Qwen3-8B, using llama.cpp's convert_lora_to_gguf.py.
It distills a frontier model's tool-calling behaviour on a market-analyst ReAct task into Qwen3-8B. Scored per single decision against the teacher, tool-selection accuracy rose from 49% (untrained base) to 90% (this adapter). Training code, harvest pipeline, and eval: https://github.com/ibrahimhka/marketanalyst-ReAct-distill
The adapter is applied on top of a Qwen3-8B base GGUF (e.g. Qwen/Qwen3-8B-GGUF):
CLI approach:
llama-cli -m qwen3-8b.gguf --lora lora-qwen3-8b.gguf (...other args)
Server mode:
llama-server -m qwen3-8b.gguf --lora lora-qwen3-8b.gguf (...other args)
For comprehensive server configuration guidance, consult the official llama.cpp repository documentation.
The original PEFT adapter (adapter_model.safetensors) is also included for use with
transformers / unsloth on top of the base model. See the
GitHub repository for the
training and inference code.