metadata
license: apache-2.0
language:
- en
pipeline_tag: text-generation
base_model:
- Qwen/Qwen2.5-3B-Instruct
tags:
- domain-finetuning
- finance
- qwen
- tgi
- vercel
FinSight LLM (Domain-FT Qwen 3B)
TL;DR: A domain-tuned finance Q&A model (Qwen 3B) for ratios, filings, and valuation topics.
Deployed via Text Generation Inference (TGI); frontend: Next.js (Vercel).
Intended use
- Educational finance Q&A, ratio explanations, simplified filings summaries.
- Not for investment advice or execution. See Limitations & Safety.
How to use
Python (Transformers)
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model_id = "willckim/domain-ft-qwen3b"
tok = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto")
prompt = "Explain PEG vs P/E with a 1-liner example."
x = tok(prompt, return_tensors="pt").to(model.device)
y = model.generate(**x, max_new_tokens=256, temperature=0.2)
print(tok.decode(y[0], skip_special_tokens=True))