coderonion
Follow
Python
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train gpt-oss, Qwen3, Llama 4, DeepSeek-R1, Gemma 3, TTS 2x faster with 70% less VRAM.
Finetune Llama 3.3, DeepSeek-R1 & Reasoning LLMs 2x faster with 70% less memory
⚡️FFPA: Extend FlashAttention-2 with Split-D, achieve ~O(1) SRAM complexity for large headdim, ~2x↑ vs SDPA EA.🎉
📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉
Democratizing Reinforcement Learning for LLMs
Effortless data labeling with AI support from Segment Anything and other awesome models.
🚀 Efficient implementations of state-of-the-art linear attention models
Simple, scalable AI model deployment on GPU clusters
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
【EMNLP 2024🔥】Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!
🚀 「大模型」3小时从0训练27M参数的视觉多模态VLM！🌏 Train a 27M-parameter VLM from scratch in just 3 hours!
OpenSeek aims to unite the global open source community to drive collaborative innovation in algorithms, data and systems to develop next-generation models that surpass DeepSeek.
OpenSeek aims to unite the global open source community to drive collaborative innovation in algorithms, data and systems to develop next-generation models that surpass DeepSeek.
label your image with Segment Anything Model or MobileSAM, support remote labeling for multiple persons。使用Segment Anything Model或MobileSAM辅助标注的工具，支持多人远程标注
Effortless AI-assisted data labeling with AI support from YOLO, Segment Anything (SAM+SAM2), MobileSAM!!
Solve Visual Understanding with Reinforced VLMs
Build multimodal language agents for fast prototype and production
Wan: Open and Advanced Large-Scale Video Generative Models
Explore the Multimodal “Aha Moment” on 2B Model
Explore the Multimodal “Aha Moment” on 2B Model
No fortress, purely open ground. OpenManus is Coming.
No fortress, purely open ground. OpenManus is Coming.
SGLang is a fast serving framework for large language models and vision language models.
High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.
High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.