InfluxDB 3 OSS is now GA. Transform, enrich, and act on time series data directly in the database. Automate critical tasks and eliminate the need to move data externally. Download now. Learn more →
Top 23 Python large-language-model Projects
-
Langflow: https://www.langflow.org/
-
InfluxDB
InfluxDB – Built for High-Performance Time Series Workloads. InfluxDB 3 OSS is now GA. Transform, enrich, and act on time series data directly in the database. Automate critical tasks and eliminate the need to move data externally. Download now.
-
gpt_academic
为GPT/GLM等LLM大语言模型提供实用化交互接口，特别优化论文阅读/润色/写作体验，模块化设计，支持自定义快捷按钮&函数插件，支持Python和C++等项目剖析&自译解功能，PDF/LaTex论文翻译&总结功能，支持并行问询多种LLM模型，支持chatglm3等本地模型。接入通义千问, deepseekcoder, 讯飞星火, 文心一言, llama2, rwkv, claude2, moss等。
- Project mention: Llama-Factory: Unified, Efficient Fine-Tuning for 100 Open LLMs | news.ycombinator.com | 2025-09-18
-
storm
An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.
Project mention: Code Explanation: "STORM: Synthesis of Topic Outlines through Retrieval and Multi-perspective Question Asking" | dev.to | 2025-03-08Note: this explanation only covers the knowledge_storm in the storm repo because it aligns with my interests.
- Project mention: 💻 Unlock RAG-Anything’s Power with Ollama on Your Machine (with Docling as Bonus) | dev.to | 2025-12-01
Modern documents increasingly contain diverse multimodal content — text, images, tables, equations, charts, and multimedia — that traditional text-focused RAG systems cannot effectively process. RAG-Anything addresses this challenge as a comprehensive All-in-One Multimodal Document Processing RAG system built on LightRAG.
-
Qwen
The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.
Project mention: Running Qwen, Nearly as Powerful as DeepSeek, on a MacBook Pro | dev.to | 2025-02-05Qwen (Qwen GitHub Repository) has been gaining attention recently as a powerful open-source large language model (LLM). I decided to give it a spin on my MacBook Pro using Ollama, a platform designed for running local LLMs. While Qwen2.5-Max boasts the highest performance, my setup could only handle the smaller Qwen2.5 (32B) model. Here's what I found!
-
-
Stream
Stream - Scalable APIs for Chat, Feeds, Moderation, & Video. Stream helps developers build engaging apps that scale to millions with performant and flexible Chat, Feeds, Moderation, and Video APIs and SDKs powered by a global edge network and enterprise-grade infrastructure.
-
langextract
A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.
langextract: A tool for extracting language information. View on GitHub
-
NeMo
A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)
git clone https://github.com/NVIDIA/NeMo.git nemo
-
For kernel-level performance tuning you can use the occupancy calculator as pointed out by jplusqualt or you can profile your kernel with Nsight compute which will give you a ton of info.
But for model-wide performance, you basically have to come up with your own calculation to estimate the FLOPs required by your model and based on that figure out how well your model is maxing out the GPU capabilities (MFU/HFU).
Here is a more in-depth example on how you might do this: https://github.com/stas00/ml-engineering/tree/master/trainin...
-
-
camel
🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org
[Enhance] tokenlimit Summarize up to the Last User Message · Issue #3371 · camel-ai/camel
-
-
GitHub Link: https://github.com/agentscope-ai/agentscope Summary: Agentscope is an agent-oriented programming library that makes it easier to build LLM applications. It's designed to be "developer-centric" with features like asynchronous execution, parallel tool calls, and real-time steering. It offers a transparent approach where prompt engineering and API invocation are fully visible and controllable. Why it's important: Agentscope, along with its related libraries like agentscope-runtime and agentscope-studio, provides a comprehensive toolkit for not only developing but also deploying and visualizing agent-based applications.
- Project mention: Lightning.ai – an enterprise managed inference platform for AI | news.ycombinator.com | 2025-10-09
After making model training simpler with PyTorch Lightning, Lightning.AI is now tackling the next bottleneck — inference. Their new managed service targets enterprises deploying LLMs and deep learning models at scale, emphasizing performance, cost-efficiency, and developer-friendly tooling.
Platform: https://lightning.ai/
-
This paper I read from here has an interesting mathematical model for reasoning based on cognitive science. https://arxiv.org/abs/2506.21734 (there is also code here https://github.com/sapientinc/HRM) I think we will see dramatic performance increases on "reasoning" problems when this is worked into existing AI architectures.
-
-
- Project mention: The AI-Native GraphDB + GraphRAG + Graph Memory Landscape & Market Catalog | dev.to | 2025-10-26
GitHub: https://github.com/neuml/txtai
-
petals
🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading
Project mention: Petals: Run large language models at home, BitTorrent‑style | news.ycombinator.com | 2025-05-27 -
-
OpenRLHF
An Easy-to-use, Scalable and High-performance RLHF Framework based on Ray (PPO & GRPO & REINFORCE++ & TIS & vLLM & Ray & Dynamic Sampling & Async Agentic RL)
-
-
SaaSHub
SaaSHub - Software Alternatives and Reviews. SaaSHub helps you find the best software and product alternatives
NOTE: The open source projects on this list are ordered by number of github stars. The number of mentions indicates repo mentiontions in the last 12 Months or since we started tracking (Dec 2020).
Python large-language-models discussion
Python large-language-models related posts
-
💻 Unlock RAG-Anything’s Power with Ollama on Your Machine (with Docling as Bonus)
-
Program-of-Thought Prompting Outperforms Chain-of-Thought by 15%
-
7 Ways to Create High-Quality Evaluation Datasets for LLMs
-
73% of AI startups are just prompt engineering
-
Brainwash Your Agent: How We Keep The Memory Clean
-
Building Custom Components in Langflow 🛠️
-
🍥 Hands-on Experience with LightRAG
- A note from our sponsor - InfluxDB www.influxdata.com | 23 Dec 2025
Index
What are some of the best open-source large-language-model projects in Python? This list will help you:
| # | Project | Stars |
|---|---|---|
| 1 | langflow | 141,437 |
| 2 | gpt_academic | 69,839 |
| 3 | LLaMA-Factory | 64,310 |
| 4 | storm | 27,730 |
| 5 | LightRAG | 26,055 |
| 6 | Qwen | 19,974 |
| 7 | Chinese-LLaMA-Alpaca | 18,967 |
| 8 | langextract | 17,474 |
| 9 | NeMo | 16,336 |
| 10 | ml-engineering | 16,071 |
| 11 | ChatGLM2-6B | 15,696 |
| 12 | camel | 15,069 |
| 13 | Megatron-LM | 14,637 |
| 14 | agentscope | 14,542 |
| 15 | litgpt | 13,034 |
| 16 | HRM | 12,150 |
| 17 | MOSS | 12,049 |
| 18 | LLMSurvey | 12,026 |
| 19 | txtai | 11,949 |
| 20 | petals | 9,844 |
| 21 | PentestGPT | 10,335 |
| 22 | OpenRLHF | 8,613 |
| 23 | optimate | 8,366 |
Sponsored
InfluxDB – Built for High-Performance Time Series Workloads
InfluxDB 3 OSS is now GA. Transform, enrich, and act on time series data directly in the database. Automate critical tasks and eliminate the need to move data externally. Download now.
www.influxdata.com