The offical implementation of DeMix (Decouple Searching from Training: Scaling Data Mixing via Model Merging for Large Language Model Pre-training)
☆46May 18, 2026Updated 5 months ago
Alternatives and similar repositories for DeMix
Users that are interested in DeMix are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- ☆30Oct 8, 2026Updated last week
- Generative models from scratch, progressively implemented in NumPy, C, and CUDA to uncover what ML frameworks hide.☆160Oct 6, 2026Updated last week
- LeWAM: End-to-End World and Action Modeling with JEPAs☆268Updated this week
- Static-first Astro framework for brand sites, Sanity CMS, and read-only Shopify retail catalogs. 0.x preview.☆51Oct 11, 2026Updated last week
- Taking the pi coding agent to Claude Code-level performance: 0.539 → 0.773 pass@1 on Terminal-Bench 2.1 with the same self-hosted 27B mod…☆186Oct 7, 2026Updated last week
- Bare Metal GPUs on DigitalOcean Gradient AI • AdPurpose-built for serious AI teams training foundational models, running large-scale inference, and pushing the boundaries of what's possible.
- Local Windows toolkit for watermarking images and converting between images and PDF.☆61Aug 13, 2026Updated 2 months ago
- A Flash, Omni info discovery radar for any topic with zero-cost.☆147Sep 4, 2026Updated last month
- Engram 是一款主打“激进速通”的高强度英语词汇记忆的桌面应用。 它的设计初衷是剥离一切非必要的视觉干扰与游戏化机制，提供一个极简、专注的终端，让学习流程保持直接、稳定和高频，以应对短期内需要极大量扩充词汇储备的备考需求（如 CET-6 突破、雅思前置准备等）。☆23Jun 10, 2026Updated 4 months ago
- AI股票量化交易大A☆49Jul 21, 2026Updated 2 months ago
- Open-source Rust TemporalStore for LLM context management, temporal features, and risk control☆212Updated this week
- A visual map for long AI conversations (ChatGPT / Claude / Gemini)☆21Jun 21, 2026Updated 3 months ago
- Local skill for turning repeated multi-step tasks into reusable workflows and SOPs.☆24Apr 3, 2026Updated 6 months ago
- A professional frontend-only Rubik's Cube & speedsolving timer. Features WCA inspection, scrambles, and stats. Hosted at ztimer.zcompany.…☆50Aug 19, 2026Updated last month
- ☆52Jun 21, 2026Updated 3 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.
- Closed-loop fault localization and checkpoint recovery for LLM Pre-training☆189Updated this week
- ☆91Updated this week
- Train your AI coders the way you'd train a vine — discover your codebase's actual patterns, then encode them as Claude Code rules, skills…☆69Updated this week
- A langGraph-powered multi-agent RAG assistant for warranty, return, and consumer decision workflow.☆120Sep 26, 2026Updated 3 weeks ago
- Let a running Claude Code session set its own native /goal — including background / non-tmux sessions — via tmux send-keys or a fail-clos…☆101Jul 8, 2026Updated 3 months ago
- Histology Spatial Topology for Omics Profiling and Inter-section Alignment☆777Updated this week
- 数据服务平台☆39Jul 14, 2026Updated 3 months ago
- 一套完全运行于本地的论文管理与 AI 智能分析系统。通过大语言模型与本地 RAG，打通“文献管理 — 智能解析 — 沉浸式阅读 — 个人笔记”的科研全工作流。☆41Sep 6, 2026Updated last month
- ☆979Jul 22, 2026Updated 2 months ago
- 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.
- Compact AI agent runtime control plane with governed tools, HITL approval, resumable execution, trace-driven evaluation, and SWE-bench-sh…☆40Aug 26, 2026Updated last month
- AI-Powered Zettelkasten Desktop Agent — local-first second brain with hybrid search, AI agents, knowledge graph, and intelligent canvas. …☆90Aug 26, 2026Updated last month
- Multi-LLM deliberation system with 6 protocols: council, debate, red-team, consensus, specialist, tournament☆131Aug 28, 2026Updated last month
- Run GPT Responses inside the Claude Code harness with protocol-correct tools, agents, streaming, and compaction.☆102Aug 11, 2026Updated 2 months ago
- Audit AI chat exports locally before sharing or migration. Find possible secrets, personal-data patterns, broken JSON/JSONL, SQLite issue…☆874Aug 25, 2026Updated last month
- 别读评论区了，让你的 Agent 去和专家吵一架。☆87Mar 21, 2026Updated 6 months ago
- 按课表记好每一节课、被点名时及时提醒的本地优先 Android 课堂助手：离线录音转写、点名提醒、课后回放搜索与可选 AI 答案☆101Updated this week
- All-in-One Multimodal Parsing Engine + Ontology-Powered, LLM Wiki-Driven AI-Ready Knowledge Engine☆1,406Oct 10, 2026Updated last week
- macOS AI 语音输入工具，支持中文与中英混合识别、AI 文本整理、上下文格式匹配和本地优先。☆291Aug 14, 2026Updated 2 months ago
- Managed Kubernetes at scale on DigitalOcean • AdDigitalOcean Kubernetes includes the control plane, bandwidth allowance, container registry, automatic updates, and more for free.
- 一个面向长视频内容理解的 Video Agent 平台。☆89Oct 7, 2026Updated last week
- [ICML 2026] Benchmarking and Evolving Reason-Reflect-Rectify for Reflective Visual Generation☆108Jun 16, 2026Updated 4 months ago
- Model authenticity / dilution detection CLI: one command, one report. Detect fake or watered-down LLM endpoints via identity, context, to…☆41Updated this week
- 《从零开始玩转 DeepSeek Harness》——DeepSeek Harness 实战指南☆153Sep 12, 2026Updated last month
- 一个适用于多任务系统+LVGL 8.x 的轻量 MVVM 支持库。☆118Sep 2, 2026Updated last month
- Rollout inference runtime for Agentic RL and multi-turn code agents☆100Sep 15, 2026Updated last month
- Agnes Harness: A pluggable agent harness for Forward Deployed Engineering (FDE). Built for trust. Made for the real world.☆777Updated this week