xie-lab-ml/Mano-Restriking-Manifold-Optimization-for-LLM-Training

The official code of "Mano: Restriking Manifold Optimization for LLM Training".

GitHub repository with 25 stars and 5 forks.

Language: Python

Topics: llm-training, optimizer, training

Open provider repository

Latest metric snapshot

2026-06-04: 25 stars and 5 forks.

Similar repositories

  1. 1. skypilot-org/skypilot

    Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

    GitHub repository with 10,113 stars and 1,094 forks.

    Trending score: 1.89; stars gained: +11; forks gained: +2.

    Language: Python

    Topics: cloud-computing, cloud-management, cost-optimization, deep-learning, distributed-training, gpu

  2. 2. ChenLiu-1996/LM-Dispersion

    [๐—œ๐—–๐— ๐—Ÿ ๐Ÿฎ๐Ÿฌ๐Ÿฎ๐Ÿฒ] Dispersion loss counteracts embedding condensation and improves generalization in small language models

    GitHub repository with 107 stars and 5 forks.

    Trending score: 0.95; stars gained: +2; forks gained: +0.

    Language: Python

    Topics: icml, icml-2026, llm, llm-training, llms, manifold-learning

  3. 3. rllm-org/rllm

    Democratizing Reinforcement Learning for LLMs

    GitHub repository with 5,608 stars and 576 forks.

    Trending score: 0.94; stars gained: +1; forks gained: +1.

    Language: Python

    Topics: agent-framework, agentic-workflow, distributed-training, llm-training, machine-learning, ml-infrastructure

  4. 4. Mattral/Composed-Mixture-of-Experts-Engine

    moe-engine is a research-grade infrastructure layer for training large Mixture-of-Experts language models at hyperscale. It is designed around one core constraint: at 10K+ GPUs, nodes die continuously. The system must keep training alive end-to-end โ€” routing correctly, checkpointing durably, and resuming without operator intervention.

    GitHub repository with 10 stars and 8 forks.

    Trending score: 0.50; stars gained: +0; forks gained: +0.

    Language: Python

    Topics: distributed-training, fault-tolerance, llm-training, machine-learning, mixture-of-experts, moe

  5. 5. Eternego-AI/eternego

    The Eternego codebase

    GitHub repository with 12 stars and 0 forks.

    Trending score: 0.05; stars gained: +0; forks gained: +0.

    Language: Python

    Topics: ai, assistant, llm, llm-agent, llm-security, llm-training

  6. 6. ludwig-ai/ludwig

    Low-code framework for building custom LLMs, neural networks, and other AI models

    GitHub repository with 11,708 stars and 1,220 forks.

    Trending score: 0.03; stars gained: +0; forks gained: +0.

    Language: Python

    Topics: computer-vision, data-centric, data-science, deep, deep-learning, deeplearning

Trending in Python

  1. 1. mvanhorn/last30days-skill

    AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary

    GitHub repository with 40,614 stars and 3,271 forks.

    Trending score: 5.82; stars gained: +1,312; forks gained: +87.

    Language: Python

  2. 2. chopratejas/headroom

    Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.

    GitHub repository with 24,986 stars and 1,636 forks.

    Trending score: 5.73; stars gained: +2,844; forks gained: +202.

    Language: Python

    Topics: agent, ai, anthropic, claude-code, compression, context-engineering

  3. 3. pewdiepie-archdaemon/odysseus

    Self-hosted AI workspace.

    GitHub repository with 69,622 stars and 8,812 forks.

    Trending score: 5.70; stars gained: +951; forks gained: +165.

    Language: Python

  4. 4. NousResearch/hermes-agent

    The agent that grows with you

    GitHub repository with 192,291 stars and 33,524 forks.

    Trending score: 5.48; stars gained: +990; forks gained: +282.

    Language: Python

    Topics: ai, ai-agent, ai-agents, anthropic, chatgpt, claude

  5. 5. safishamsi/graphify

    AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph.

    GitHub repository with 66,406 stars and 6,716 forks.

    Trending score: 5.25; stars gained: +1,314; forks gained: +109.

    Language: Python

    Topics: claude-code, graphrag, knowledge-graph, codex, openclaw, skills

  6. 6. hugohe3/ppt-master

    AI generates a real, editable PowerPoint from any document โ€” native shapes & animations, speaker notes voiced as audio narration, and the option to follow your own .pptx template, not slide images ยท by Hugo He

    GitHub repository with 27,093 stars and 2,416 forks.

    Trending score: 5.10; stars gained: +903; forks gained: +61.

    Language: Python

    Topics: ai-agent, powerpoint, pptx, presentation, office, slides

Trending topic: llm-training

  1. 1. openlake-project/openlake

    High performance storage system for LLM Inference and GPU Training. Feed your GPUs at blazing fast speeds

    GitHub repository with 1,339 stars and 198 forks.

    Trending score: 2.76; stars gained: +12; forks gained: +4.

    Language: Rust

    Topics: gpu, high-performance, rdma, rust, storage, throughput

  2. 2. chrisliu298/awesome-on-policy-distillation

    A curated collection of papers, technical reports, frameworks, and tools for on-policy distillation (OPD) of large language models

    GitHub repository with 320 stars and 7 forks.

    Trending score: 1.93; stars gained: +11; forks gained: +1.

    Topics: on-policy-distillation, awesome, llm, rl, opd, awesome-list

  3. 3. skypilot-org/skypilot

    Run, manage, and scale AI workloads on any AI infrastructure. Use one system to access & manage all AI compute (Kubernetes, Slurm, 20+ clouds, on-prem).

    GitHub repository with 10,113 stars and 1,094 forks.

    Trending score: 1.89; stars gained: +11; forks gained: +2.

    Language: Python

    Topics: cloud-computing, cloud-management, cost-optimization, deep-learning, distributed-training, gpu

  4. 4. ChenLiu-1996/LM-Dispersion

    [๐—œ๐—–๐— ๐—Ÿ ๐Ÿฎ๐Ÿฌ๐Ÿฎ๐Ÿฒ] Dispersion loss counteracts embedding condensation and improves generalization in small language models

    GitHub repository with 107 stars and 5 forks.

    Trending score: 0.95; stars gained: +2; forks gained: +0.

    Language: Python

    Topics: icml, icml-2026, llm, llm-training, llms, manifold-learning

  5. 5. rllm-org/rllm

    Democratizing Reinforcement Learning for LLMs

    GitHub repository with 5,608 stars and 576 forks.

    Trending score: 0.94; stars gained: +1; forks gained: +1.

    Language: Python

    Topics: agent-framework, agentic-workflow, distributed-training, llm-training, machine-learning, ml-infrastructure

  6. 6. bd4sur/Nano

    ็”ตๅญ้นฆ้น‰ / Toy Language Model

    GitHub repository with 291 stars and 13 forks.

    Trending score: 0.52; stars gained: +1; forks gained: +0.

    Language: C

    Topics: ai, gpt, llm, llm-inference, llm-training, on-device