Transformer: open-source projects on GitHub
35 repositories on Transformer, picked by the REDDYX AI agent. Each has a summary, key features and a breakdown. Sorted by stars.
// RELATED TOPICS
🤗 Transformers
A framework for defining machine learning models in text, vision, audio, and multimodal models, for inference and training.
★ 160,482 · PythonHigh-performance LLM engine
A high-performance and memory-efficient engine for inference and serving large language models.
★ 79,783 · Pythonwhisper.cpp: C/C++ port of Whisper
High-performance port of the OpenAI Whisper speech recognition model in C/C++ for local and cross-platform use.
★ 49,828 · C++SGLang: LLM Framework
SGLang is a high-performance framework for serving large language and multimodal models.
★ 27,854 · PythonFAST SPEECH TRANSCRIPTION
Faster Whisper — an accelerated version of the Whisper model for speech recognition, built on CTranslate2. It supports quantized models and provides real‑time transcripti
★ 26,758 · PythonUncensored LLM Automation
The project eliminates automatic censorship in language models, making responses free and unrestricted.
★ 20,881 · PythonFreeToken: Local Model Run
FreeToken lets you run large AI models on your PC, delivering data‑center‑class speed and efficiency without needing cloud resources.
★ 17,197 · PythonMNN: FAST AI ON DEVICE
Lightweight and lightning-fast inference engine from Alibaba for on-device LLM and Edge AI.
★ 15,386 · C++KIMI K3: 2.78 Trillion Parameters
Running the Kimi K3 model with 2.78 trillion parameters on a single CPU, using only 8.24 GB of RAM without BLAS, frameworks, or GPU.
★ 12,109 · CAI FRAME FOR DEVICES
This project offers low-bit quant kernels and runtime for consumer devices on Android/ARM. It solves edge-ai tasks with minimal resources.
★ 6,192 · C++Efficient AI for Mobile Devices
An AI engine with minimal latency for smartphones and wearable devices.
★ 4,949 · CEfficient Multimodal Inference
Framework for efficient inference of multimodal models. Unifies text, images, and audio in a single generation system.
★ 4,863 · PythonDIY-LLM: FULL LLM COURSE WITH CODE
The diy-llm project offers six step‑by‑step Jupyter notebooks covering data pre‑training, tokenization, Transformer, MoE, distributed training, scaling laws, inference an
★ 2,960 · Jupyter NotebookAwesome LLM Long Context - Top Papers
A curated selection of must-read articles and blogs on long context in LLMs: new methods, reviews, and practices.
★ 2,192VLLM-ASCEND: ASCEND PLUGIN ON GITHUB
Community plugin for vLLM on Ascend hardware from the community, written in C++. Gaining stars — already 2156!
★ 2,156 · C++GRAPHS VS FRAUD
Curated list of articles on fraud detection using graphs and transformers.
★ 1,907VIBETHINKER: 1.5B Model
Tiny model with powerful logical reasoning — Diversity-Driven Optimization awakens the mind of a large LLM. 577 ★ already!
★ 1,589 · PythonPARALLAX: BUILD AI CLUSTER
Parallax lets you quickly deploy a distributed inference service on any infrastructure.
★ 1,301 · PythonMULTIMODAL AI FOR CONTENT
UForm — compact MULTIMODAL AI for understanding and generating content. Processes multilingual texts, images and video, working 5x faster than OpenAI CLIP and LLaVA.
★ 1,253 · PythonAI-HANDS-ON: LEARN AI FROM SCRATCH
A collection of notebooks and files for learning AI from scratch. 1137 stars on GitHub.
★ 1,137 · Jupyter NotebookSPARK-X2.6: AGENTS ON DEVICE
Open Spark-x2.5 model series focused on agentic capabilities. Runs locally on device without cloud. 814 stars on GitHub.
★ 911MULTILSPY — PYTHON LSP CLIENT
multilspy is a Python LSP-client library that simplifies interaction with language servers for IDEs and code analysis.
★ 615 · PythonNext-Gen Transformer Framework
Open-source framework for next-gen Transformers with compute-matched benchmarking and autonomous architecture research platform.
★ 412 · PythonZINC - LOCAL LLM ON GPU
Zinc is an LLM inference engine in Zig that runs on AMD GPU and Apple Silicon without external dependencies.
★ 395 · ZigBUILD LLM IN SECONDS
CLI tool creates ready-to-train PyTorch projects for LLM in seconds. Analog of create‑next‑app, but for language models. Written in TypeScript, already has 362 stars.
★ 373 · TypeScriptawesome-AI4MolConformation-MD - AI MOLECULE GENERATION
A collection of conformations of small molecules, RNA, peptides, proteins, and enzymes obtained using generative AI and molecular dynamics.
★ 307TITLE - HONGBOMIAO LAB
Personal R&D laboratory on Python with 298 stars, uniting developers to share knowledge and grow together.
★ 300 · PythonCEVAHIR AI OPEN AI ENGINE
Full-featured open-source engine for training tokenizers, transformers, and cognitive reasoning, ready for chat flow.
★ 293 · PythonMONITOR-GUIDED DECODING FOR CODE LLM
NeurIPS 2023 artifact: Monitor-Guided Decoding of code LMs with static analysis of repository context. Includes multispy — Python LSP client for language server integrati
★ 279 · PythonLight fine-tuning of LLM assistants
Zero-friction fine-tuning of any LLMs for Claude Code, Gemini CLI, and ACP agents: from environment setup to training (LoRA, SFT, DPO, GRPO) with NVIDIA and Apple Silicon
★ 237 · PythonTRANSFORMER-TRICKS - ACCELERATING TRANSFORMERS
A collection of techniques and utilities for accelerating transformers: attention optimization, quantization, and caching of computations.
★ 236 · TeXVLLM INFERENCE: LOOPED TRANSFORMERS
vllm-rlt — vLLM-style inference engine for looped transformers. Implements loop-level scheduling and execution via composable stages. 207 stars on GitHub.
★ 227 · PythonHands-On LLM Full-Stack Course
Comprehensive tutorial implementing LLMs from scratch: builds Transformer in PyTorch, reproduces Pretrain/SFT/LoRA/DPO/GRPO on single GPU, covers PEFT/Agent/RAG with real
★ 124 · PythonLLM course from scratch
Practical course on 23 Jupyter notebooks where you build tokenizers, attention, MoE, RLHF, inference, evaluation, and distillation of modern LLMs in PyTorch.
★ 53 · Jupyter NotebookDVLT on CUDA without dependencies
The project ports NVIDIA DVLT to pure CUDA/C++, allowing images to be fed in and a 3D point cloud together with camera poses to be obtained. Everything is packed into a s
★ 31 · CudaNew projects on this topic in Telegram
Telegram →