BizFinBench
A Business-Driven Real-World Financial Benchmark for Evaluating LLMs
GraphCanon updated 3w · GitHub synced 3w
Decision brief
BizFinBench is a finance-specific benchmark for evaluating large language models in real-world business settings.
Good fit when
- For teams专注于金融行业，需要评估其模型在实际业务场景中的表现时。
- 如果需要使用具体的财务基准来衡量大型语言模型的性能，并且希望该基准能反映真实世界的商业情景，则适合采用BizFinBench。
Avoid when
- 不要在不需要针对财务管理特定挑战进行测试的情况下使用BizFinBench，例如，在评估通用自然语言处理任务时。
- 如果项目主要关注的是非金融行业的应用，并且所使用的场景与实际的商业交易无直接关联，则不宜采用BizFinBench进行测评。
- Requirements:
- 需要安装所需的Python库以运行评估：pip install -r requirements.txt; 环境变量设置包括模型路径、远程模型URL、模型名称以及其他相关参数，如使用API进行测试时的API key等。; 必须确保遵守相关的使用和许可政策，这可能涉及研究使用的限制及其他第三方协议条款。
Observed Jul 15, 2026 · Source: enrich:decision_facts
Verify the decision
Maintenance and security
Full trust report- Maintenance
- Steady (88d since push)
- As of 3w
- Provenance
- Not a fork · Organization account
- As of 3w
- Security (OSV)
- 90 low (90 low)
- As of 1mo
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
pip install BizFinBenchSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
BizFinBench is designed as a finance-oriented benchmark to evaluate the performance of large language models in real-world business scenarios.
Capability facts
- Languages
- python
Source: github.language · Jul 29, 2026
Categories
Compatibility
Sourced claims from the README excerpt - not unsourced marketing copy.
Source: README excerpt (regex_v1, Jul 29, 2026)
pip install -r requirements.txtSource link
Source: README excerpt (regex_v1, Jul 29, 2026)
### Quick Start – Evaluate external apis (e.g., chatgpt)Source link
Tags
README
Install requirements
pip install -r requirements.txt
Quick Start – Evaluate a Local Model
export MODEL_PATH=model/Qwen2.5-0.5B # Path to the model to be evaluated
export REMOTE_MODEL_PORT=16668
export REMOTE_MODEL_URL=http://127.0.0.1:${REMOTE_MODEL_PORT}/model
export MODEL_NAME=Qwen2.5-0.5B
export PROMPT_TYPE=chat_template # Hithink llama3 llama2 none qwen chat_template; chat_template is recommended
---
### Quick Start – Evaluate a Local Model and Score with a Judge Model
```sh
export MODEL_PATH=model/Qwen2.5-0.5B # Path to the model to be evaluated
export REMOTE_MODEL_PORT=16668
export REMOTE_MODEL_URL=http://127.0.0.1:${REMOTE_MODEL_PORT}/model
export MODEL_NAME=Qwen2.5-0.5B
export PROMPT_TYPE=chat_template # llama3 llama2 none qwen chat_template; chat_template is recommended
---
### Quick Start – Evaluate external apis (e.g., chatgpt)
```sh
export API_NAME=chatgpt # The api name, currently support chatgpt
export API_KEY=xxx # Your api key
export MODEL_NAME=gpt-4.1
---
## 📄 License
**Usage and License Notices**: The data and code are intended and licensed for research use only.
License: Attribution-NonCommercial 4.0 International It should abide by the policy of OpenAI: https://openai.com/policies/terms-of-use
For agents
This page has a .md twin and JSON over the API.