Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
LLM-based ontological extraction tools, including SPIRES
| Date | Stars |
|---|---|
| 2026-07-31 | 953 |
| 2026-08-02 | 954 |
| 2026-08-06 | 954 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# OntoGPT
[](https://zenodo.org/badge/latestdoi/13996/monarch-initiative/ontogpt)
## Introduction
_OntoGPT_ is a Python package for extracting structured information from text with large language models (LLMs), _instruction prompts_, and ontology-based grounding.
[For more details, please see the full documentation.](https://monarch-initiative.github.io/ontogpt/)
## Quick Start
OntoGPT runs on the command line, though there's also a minimal web app interface (see `Web Application` section below).
1. Ensure you have Python 3.10 or greater installed.
2. Install with `pip`:
```bash
pip install ontogpt
```
3. Set your OpenAI API key:
```bash
runoak set-apikey -e openai <your openai api key>
```
4. See the list of all OntoGPT commands:
```bash
ontogpt --help
```
5. Try a simple example of information extraction:
```bash
echo "One treatment for high blood pressure is carvedilol." > example.txt
ontogpt extract -i example.txt -t drug
```
OntoGPT will retrieve the necessary ontologies and output results to the command line. Your output will provide all extracted objects under the heading `extracted_object`.
## Web Application
There is a bare bones web application for running OntoGPT and viewing results.
First, install the required dependencies with `pip` by running the following command:
```bash
pip install ontogpt[web]
```
Then run this command to start the web application:
```bash
web-ontogpt
```
NOTE: We do not recommend hosting this webapp publicly without authentication.
## Model APIs
OntoGPT uses [LiteLLM](https://docs.litellm.ai/docs/) to interface with LLMs.
This means OntoGPT can work with a much broader range of providers than just OpenAI. If a provider and model are supported by the installed LiteLLM version, they will generally work in OntoGPT as well. This includes OpenAI, Azure OpenAI, Anthropic, Mistral, Groq, Cohere, Vertex AI, Replicate, and many others.
The model name to use may be found from the command `ontogpt list-models` - use the name in the first column with the `--model` option. In most cases, the most reliable form is a provider-qualified LiteLLM model name such as `openai/gpt-4o`, `anthropic/claude-3-5-sonnet`, `groq/llama-3.1-8b-instant`, or `mistral/mistral-large-latest`.
Credential handling now follows LiteLLM first. Standard LiteLLM environment variables such as `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `GROQ_API_KEY`, `MISTRAL_API_KEY`, `AZURE_API_KEY`, `AZURE_API_BASE`, and `AZURE_API_VERSION` are supported directly. For backward compatibility, OntoGPT also checks Oaklib credentials created with `runoak set-apikey` and passes them through to LiteLLM when the corresponding provider settings are missing.
Examples:
```bash
runoak set-apikey -e openai <your openai api key>
runoak set-apikey -e anthropic-key <your anthropic api key>
runoak set-apikey -e mistral-key <your mistral api key>
runoak set-apikey -e groq-key <your groq api key>
```
Some endpoints, such as Azure OpenAI, require additional details. These may be set similarly:
```bash
runoak set-apikey -e azure-key <your azure api key>
runoak set-apikey -e azure-base <your azure endpoint url>
runoak set-apikey -e azure-version <your azure api version, e.g. "2023-05-15">
```
These details may also be set as environment variables as follows:
```bash
export AZURE_API_KEY="my-azure-api-key"
export AZURE_API_BASE="https://example-endpoint.openai.azure.com"
export AZURE_API_VERSION="2023-05-15"
```
If the provider is not encoded in the model name, use `--model-provider` to specify it explicitly. This is most common for OpenAI-compatible proxy endpoints.
For the current list of supported providers, model naming rules, and credential environment variables, see the LiteLLM docs:
* <https://docs.litellm.ai/docs/provExcerpt of 6,885 characters
Read on GitHub
1.7k
200
46
46
16
16
7
Yaroslav Halchenko · Dartmouth College, @dandi, @Debian, @DataLad, @neurodebian, @PyMVPA, @fail2ban
6
6
3
3
3
2
2
2
2
2
1
1
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:43e5506e40a06270, topic:nlp, topic:named-entity-recognition, topic:information-extraction
matched fp:43e5506e40a06270, topic:large-language-models, topic:llm