☆709Aug 31, 2026Updated 2 months ago
Alternatives and similar repositories for MegaTrain
Users that are interested in MegaTrain are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆14Aug 14, 2025Updated last year
- ☆13Feb 5, 2026Updated 8 months ago
- slime is an LLM training framework for RL Scaling.☆8,762Updated this week
- COS-PLAY: Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Game Play☆31Jul 11, 2026Updated 3 months ago
- 🚀 Efficient implementations for emerging model architectures☆6,004Updated this week
- Deploy to Railway using AI coding agents - Free Credits Offer • AdUse Claude Code, Codex, OpenCode, and more. Autonomous software development now has the infrastructure to match with Railway.
- NanoGPT (124M) in 90 seconds☆5,999Sep 28, 2026Updated last month
- Minimalistic large language model 3D-parallelism training☆2,850Oct 8, 2026Updated 3 weeks ago
- CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.☆560Updated this week
- Very simple regulation can mitigate reconstruction-generation gap.☆17Oct 1, 2026Updated 3 weeks ago
- A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.☆5,471May 17, 2026Updated 5 months ago
- verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework☆24,077Updated this week
- Megatron-LM adapted for Puro-2B☆103Aug 30, 2026Updated 2 months ago
- TokenSpeed is a speed-of-light LLM inference engine.☆2,824Updated this week
- Code for COLM '26 paper 'Loop, Think, & Generalize: Implicit Reasoning in Recurrent-Depth Transformers'☆108Sep 3, 2026Updated last month
- Deploy on Railway without the complexity - Free Credits Offer • AdConnect your repo and Railway handles the rest with instant previews. Quickly provision container image services, databases, and storage volumes.
- Reconstructing Big Tech T2I/T2V captioning pipelines for production and research☆45Aug 5, 2026Updated 2 months ago
- ☆22May 21, 2025Updated last year
- Multimodal Video Generation Model, research, unified models, applications collection.☆13Aug 21, 2026Updated 2 months ago
- high-performance linear attention kernel library built on TileLang☆744Updated this week
- Nano vLLM☆16,211Apr 26, 2026Updated 6 months ago
- ☆53May 20, 2025Updated last year
- Kernels, of the mega variety :)☆874May 26, 2026Updated 5 months ago
- An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Asy…☆10,130Oct 5, 2026Updated 3 weeks ago
- 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"☆1,026Feb 5, 2026Updated 8 months ago
- AI Agents on DigitalOcean Gradient AI Platform • AdBuild production-ready AI agents using customizable tools or access multiple LLMs through a single endpoint. Create custom knowledge bases or connect external data.
- Extending the Context of Pretrained LLMs by Dropping Their Positional Embedding☆225Jan 12, 2026Updated 9 months ago
- BEAVER: A Training-Free Hierarchical Prompt Compression Method via Structure-Aware Page Selection☆17Mar 30, 2026Updated 7 months ago
- ☆24Dec 12, 2025Updated 10 months ago
- Tile primitives for speedy kernels☆3,786Sep 12, 2026Updated last month
- Accelerating MoE with IO and Tile-aware Optimizations☆785Aug 29, 2026Updated 2 months ago
- An hardware-aware Efficient Implementation for "Mixture-of-Depths Attention".☆279May 6, 2026Updated 5 months ago
- [NeurIPS2026] Fast and memory-efficient exact kmeans☆742Aug 28, 2026Updated 2 months ago
- ☆1,320May 20, 2026Updated 5 months ago
- Muon is an optimizer for hidden layers in neural networks☆2,919May 24, 2026Updated 5 months ago
- Managed Database hosting by DigitalOcean • AdPostgreSQL, MySQL, MongoDB, Kafka, Valkey, and OpenSearch available. Automatically scale up storage and focus on building your apps.
- FlashInfer: Kernel Library for LLM Serving☆6,821Updated this week
- Efficient Triton Kernels for LLM Training☆6,678Updated this week
- Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.☆3,242Updated this week
- Triton kernels and PyTorch ops for Block Attention Residuals (AttnRes)☆91May 29, 2026Updated 5 months ago
- A kernel library written in tilelang☆2,019Oct 14, 2026Updated 2 weeks ago
- Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.☆1,614Oct 23, 2026Updated last week
- ☆61Dec 19, 2025Updated 10 months ago