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 LLM Projects
-
Project mention: The Real AI Startup Stack: $33M Valuations, $1.2K OpenAI Bills | dev.to | 2025-11-09
LangChain GitHub The prompt orchestration library every “AI platform” seems to use.
-
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.
-
Project mention: Windows-Use: an AI agent that interacts with Windows at GUI layer | news.ycombinator.com | 2025-09-08
-
ragflow
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
Project mention: The AI-Native GraphDB + GraphRAG + Graph Memory Landscape & Market Catalog | dev.to | 2025-10-26 -
If you're looking for open source agents, which can run locally, in Docker, or in the cloud, and which have a consistent track record of acing benchmark scores like SWE-bench, check out https://github.com/All-Hands-AI/OpenHands
We're about to release our Agent SDK (https://github.com/All-Hands-AI/agent-sdk/) which provides devs with all the nuts and bolts you need to define custom prompts, tools, security profiles, and multi-agent interfaces
-
Project mention: DeepSeek-OCR: When a Picture Is Actually Worth 10 Fewer Tokens | dev.to | 2025-10-26
One gotcha: if you're using vLLM, you'll need the 0.8.5 wheel for CUDA 11.8. Download it from vLLM releases before installing.
-
Project mention: Llama-Factory: Unified, Efficient Fine-Tuning for 100 Open LLMs | news.ycombinator.com | 2025-09-18
-
MetaGPT
🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming
Project mention: Backlog.md – CLI that auto-generates task files (took my Claude success to 95 %) | news.ycombinator.com | 2025-07-06 -
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.
-
unsloth
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.
-
Project mention: How to Build a RAG Solution with Llama Index, ChromaDB, and Ollama | dev.to | 2025-11-04
Step 2: Set up LlamaIndex and Chroma DB
-
mem0
Universal memory layer for AI Agents; Announcing OpenMemory MCP - local and secure memory management.
-
chatgpt-on-wechat
基于大模型搭建的聊天机器人，同时支持 微信公众号、企业微信应用、飞书、钉钉 等接入，可选择ChatGPT/Claude/DeepSeek/文心一言/讯飞星火/通义千问/ Gemini/GLM-4/Kimi/LinkAI，能处理文本、语音和图片，访问操作系统和互联网，支持基于自有知识库进行定制企业智能客服。
-
quivr
Opiniated RAG for integrating GenAI in your apps 🧠 Focus on your product rather than the RAG. Easy integration in existing products with customisation! Any LLM: GPT4, Groq, Llama. Any Vectorstore: PGVector, Faiss. Any Files. Anyway you want.
-
-
khoj
Your AI second brain. Self-hostable. Get answers from the web or your docs. Build custom agents, schedule automations, do deep research. Turn any online or local LLM into your personal, autonomous AI (gpt, claude, gemini, llama, qwen, mistral). Get started - free.
-
litellm
Call any LLM API with cost tracking, guardrails, logging and load balancing. 1.8k+ models, 80+ providers, 50+ endpoints (unified + native format). Available as a Python SDK or Proxy Server (AI Gateway).
[BerriAI/litellm]: LiteLLM - A simple library to call any LLM API
-
Project mention: The AI-Native GraphDB + GraphRAG + Graph Memory Landscape & Market Catalog | dev.to | 2025-10-26
URL: https://microsoft.github.io/graphrag/ and https://github.com/microsoft/graphrag and https://github.com/Azure-Samples/graphrag-accelerator
-
agenticSeek
Fully Local Manus AI. No APIs, No $200 monthly bills. Enjoy an autonomous agent that thinks, browses the web, and code for the sole cost of electricity. 🔔 Official updates only via twitter @Martin993886460 (Beware of fake account)
Project mention: A Step-By-Step Guide to Running AgenticSeek Locally: No API Needed | dev.to | 2025-05-08git clone --depth 1 https://github.com/Fosowl/agenticSeek.git
-
LightRAG examples: https://github.com/HKUDS/LightRAG/tree/main/examples
-
pandas-ai
Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.
-
Project mention: How Attention Sinks Keep Language Models Stable | news.ycombinator.com | 2025-08-08
I found a fairly large improvement in my toy transformer model where I added a "global" token akin to the CLS token in ViT.
Another approach I've seen is the "Diff transformer" from MS Research (https://github.com/microsoft/unilm/tree/master/Diff-Transfor...).
-
-
It depends on what you mean by "this." MLC's catch is that you need to define/compile models for it with TVM. Here is the list of supported model architectures: https://github.com/mlc-ai/mlc-llm/blob/main/python/mlc_llm/m...
llama.cpp has a much bigger supported model list, as does vLLM and of course PyTorch/HF transformers covers everything else, all of which work w/ ROCm on RDNA3 w/o too much fuss these days.
For inference, the biggest caveat is that Flash Attention is only an aotriton implementation, which besides being less performant sometimes, also doesn't support SWA. For CDNA there is a better CK-based version of FA, but CK doesn't not have RDNA support. There are a couple people at AMD apparently working on native FlexAttention, os I guess we'll how that turns out.
(Note the recent SemiAccurate piece was on training, which I'd agree is in a much worse state (I have personal experience with it being often broken for even the simplest distributed training runs). Funnily enough, if you're running simple fine tunes on a single RDNA3 card, you'll probably have a better time. OOTB, a 7900 XTX will train at about the same speed as an RTX 3090 (4090s blow both of those away, but you'll probably want more cards and VRAM of just move to H100s).
-
vanna
🤖 Chat with your SQL database 📊. Accurate Text-to-SQL Generation via LLMs using Agentic Retrieval 🔄.
Project mention: Beyond the Diff: How Deep Context Analysis Caught a Critical Bug in a 20K-Star Open Source Project | dev.to | 2025-10-20A developer submitted PR #951 to Vanna.ai, a popular open-source text-to-SQL tool with 20,000+ stars. The change added Databricks integration—156 lines of well-documented code supporting two connection engines (SQL warehouse and ODBC).
-
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 LLM discussion
Python LLM related posts
-
Show HN: qqqa – a fast, stateless LLM-powered assistant for your shell
-
How to Build a RAG Solution with Llama Index, ChromaDB, and Ollama
-
Show HN: AI Agent for a mobile robot in the real world
-
Tongyi DeepResearch – open-source 30B MoE Model that rivals OpenAI DeepResearch
-
Show HN: Hephaestus – Autonomous Multi-Agent Orchestration Framework
-
Ask HN: Who uses open LLMs and coding assistants locally? Share setup and laptop
-
OpenAI rejects 1,200-line community PR for Google's A2A agent protocol
-
A note from our sponsor - InfluxDB
www.influxdata.com | 16 Nov 2025
Index
What are some of the best open-source LLM projects in Python? This list will help you:
| # | Project | Stars |
|---|---|---|
| 1 | langchain | 119,630 |
| 2 | browser-use | 72,415 |
| 3 | ragflow | 67,441 |
| 4 | OpenHands | 64,958 |
| 5 | vllm | 62,592 |
| 6 | LLaMA-Factory | 62,169 |
| 7 | MetaGPT | 59,491 |
| 8 | unsloth | 48,261 |
| 9 | llama_index | 45,183 |
| 10 | mem0 | 42,849 |
| 11 | chatgpt-on-wechat | 39,676 |
| 12 | quivr | 38,596 |
| 13 | ChatTTS | 38,144 |
| 14 | khoj | 31,564 |
| 15 | litellm | 31,037 |
| 16 | graphrag | 29,114 |
| 17 | agenticSeek | 23,656 |
| 18 | LightRAG | 22,597 |
| 19 | pandas-ai | 22,534 |
| 20 | unilm | 21,827 |
| 21 | Scrapegraph-ai | 21,751 |
| 22 | mlc-llm | 21,614 |
| 23 | vanna | 21,588 |
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