# MetaGLM/FinGLM

FinGLM: 致力于构建一个开放的、公益的、持久的金融大模型项目，利用开源开放来促进「AI+金融」。

Repository: https://github.com/MetaGLM/FinGLM
Canonical: https://ross.abutalabs.com/products/finglm
Language: HTML
License Family: other
Topics: chatglm, finacial, gpt, llama, llm
Last push: 2024-05-08T13:46:36+00:00

## Health v2 (maintenance only)
Score: 27/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 76
- inputs: {"age_days": 1073, "days_push": 847, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2258, forks 313 (observed 2026-08-28T04:06:31.562027+00:00)

## What it is
FinGLM is an open, community-driven financial LLM project centered on a dialog-based question-answering system that analyzes Chinese listed-company annual reports. It bundles an end-to-end pipeline (PDF-to-text parsing, structuring data into SQL/MongoDB/Elasticsearch, ChatGLM fine-tuning with P-Tuning v2 and LoRA, and retrieval-augmented answer generation) together with a 70 GB annual-report dataset, 10,000 annotated evaluation samples, and multiple competition-winning solutions plus course materials.

## Use cases
- answer questions about a company's annual report with an LLM
- build a financial question-answering chatbot over company filings
- fine-tune ChatGLM on financial documents
- find an open dataset of Chinese listed-company annual reports
- extract and structure financial data from PDF annual reports
- learn how to apply large language models to financial analysis
- evaluate LLMs on financial QA using annotated benchmark data

## When to choose
- You want an end-to-end reference implementation for QA over Chinese listed-company annual reports
- You need the open 70 GB annual-report corpus and 10,000 human-annotated evaluation samples
- You want to compare multiple competition-winning fine-tuning and RAG approaches side by side

## When to avoid
- You need production-grade software with clear licensing - the repository has no license, so reuse rights are uncertain
- You need English-language filings or non-Chinese financial documents
- You want a turnkey, actively maintained financial assistant rather than competition code and learning materials

## Facets
- artifact type: application
- maturity: maintenance
- function: rag, llm-training, llm-inference, chatbot, nlp, machine-learning, pdf, database, search-engine, prompt-engineering
- domain: artificial-intelligence, large-language-models, fintech, machine-learning, data-science
- platform: python, cross-platform
- tags: financial-analysis, annual-reports, chatglm, fine-tuning, question-answering, document-qa, financial-dataset, chinese, lora, ptuning, competition-solutions, finance, natural-language-processing, gpu

## Member repositories
- MetaGLM/FinGLM (main) score 27

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:31.562027+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:43:38.245956+00:00, confidence not recorded.
  - readme: https://github.com/MetaGLM/FinGLM (fetched 2026-08-28T04:06:31.562027+00:00, sha c88ac8e0f206)
- Data as of 2026-08-30T08:39:29.467469+00:00.