facebookresearch / segment-anything

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

NandhaKishorM / laya

Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.

facebookresearch / sam2

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

facebookresearch / sam3

The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

checkpoint-restore / criu

Checkpoint/Restore tool

BernardoGiordano / Checkpoint

Fast and simple homebrew save management framework for 3DS and Switch.

Neroued / ninfer

High-performance single-GPU inference for selected model checkpoints and GPUs.

ThousandBirdsInc / chidori

The agent framework where every run is durable, replayable, and resumable by default.

letta-ai / agent-file

Agent File (.af): An open file format for serializing stateful AI agents with persistent memory and behavior. Share, checkpoint, and version control agents across compatible frameworks.

vstorm-co / pydantic-deepagents

Open-source, self-hosted Claude Code - a terminal AI assistant and the Python framework behind it. Tool-calling, sandboxed execution, multi-agent teams, skills, checkpoints, unlimited context - on Pydantic AI, any model.

Ephemeral-AI-Lab / layerfs

LayerFS gives every agent an isolated, disposable filesystem fork without copying the shared base. Useful states become durable, deduplicated checkpoints with workspace-scoped tool history—ready to branch, rewind, or reuse across parallel development, environment experiments, and MCTS-style rollouts.

shuxiachai / academic-commercialization-agent

AI multi-agent tool that turns research papers into cited commercialization and tech-transfer assessments: TRL, patent landscape, market and risks. CrewAI + FastAPI, with a reproducible benchmark.

kacper-daftcode / vLLM-Moet

A vLLM patch + hand‑written SM120 SASS kernels: 2‑bit MoE experts + an FP4 "delta" cache that recovers precision — matching the official (NV)FP4 checkpoint's quality on consumer Blackwell cards

google / orbax

Orbax provides common checkpointing and persistence utilities for JAX users

ancwrd1 / snx-rs

Open Source Client For Check Point VPN Tunnels

yaoyao-jpg / PhiZero

Official implementation of PhiZero: A World Model Built Around Physical Language. Code and checkpoints coming soon.

geyan21 / flex-pi

Flex-π: A multi-stream world-action model with compute flexibility: one checkpoint that deploys as a VLA, a full world model, or anything in between.

z-liu-xiugou / research-workbench-assistant

Local-first research memory for Codex: checkpoints, progress ledgers, and Markdown/JSON paper notes. Early alpha.

LMResiliency / lm-resiliency

Closed-loop fault localization and checkpoint recovery for LLM Pre-training

razr001 / langgraph-vew

LangGraph 可视化调试器 | A visual debugger for LangGraph — interactive graph canvases, streaming runs, step-through debugging, and checkpoint editing in the browser.

SiriLee / dsh-rewind

DSH 插件：便捷无感的同窗口内对话回退，从不新建分支；自带轻量工作区备份，可一并还原文件（Claude Code /rewind 语义）。 · DSH plugin: effortless in-window conversation rewind — never forking a new session; ships a lightweight workspace backup that restores files together with the rewind (Claude Code /rewind semantics).

wildminder / awesome-qwen-image

Qwen-Image 2.1. Checkpoints, quants, prompt engines, LoRAs, and tooling

r1c7 / CluxMate

AI coding agent with one Python core and three front-ends — headless CLI, Textual TUI, and an Electron desktop. Works with any OpenAI-compatible API, with risk-tiered permissions, event-sourced replayable sessions, and a fail-closed OS-level sandbox.

Buffalo2024 / Rulora

Pluggable controls for reliable AI agents: validated outputs, bounded loops and frozen collective decisions.

d2checkpoint-com / D2SteamFix

D2SteamFix is a launch wrapper for Destiny 2 intended to help alleviate frame pacing and stuttering issues for Steam users.

Shanghua-Gao / RSI-Jev

Typed-decision models (noul / choice / score) trained by a self-improving loop of AI agents — checkpoints, the code that produced them, and every version that failed.

limenagent / prodagent

A minimal, readable agent-framework kernel — understand how LangGraph/ADK work under the hood. 1800行代码的 agent 框架内核，理解 LangGraph/ADK 底层的工作原理

jianglongye / dexterous-astra

Dexterous pen spinning and Rubik's cube: sim environments, zero-shot scripts, trained checkpoints

malevrigns / atlas-agent-control-plane

AtlasAgent - an auditable AI agent control plane: evidence-backed memory, governed tool runtime, checkpoint DAG recovery, and a 55-chapter engineering tutorial. FastAPI / Next.js PWA / Textual TUI

XhinLiang / awsl

Codex-verified, Claude-compatible durable local runtime for Claude Code Workflow JS, with resume, checkpoints, budgets, and Git worktrees

InkMind-AI / Dyn-3D

TempoVista is the official implementation of Dyn-3D, a benchmark and training pipeline for 3D spatial reasoning in vision-language models. The release contains frame-selection tools, SFT and Kinematic-GSPO training code, benchmark evaluation scripts, Dyn-3D annotations, and checkpoints for Qwen3-VL and InternVL3.5.

MaxwellCCC / hangzhou-mahjong

Hangzhou Mahjong training and evaluation framework with public-information Transformers, paired branch reinforcement learning, and the r2600 checkpoint.

OpSafari / laborix

Autonomous experiment-pipeline orchestrator: versioned research-plan DSL with dependency DAGs, deterministic checkpoint/resumable executor, tamper-evident provenance ledger, UCB1/Thompson schedulers with measured regret against synthetic oracles, ablation bookkeeping, offline report synthesis. NumPy core, CPU-only torch extra.

MalganisWu / red-bean-qing-tuan

可嵌入的 TypeScript Agent Runtime —— model→tool 循环、checkpoint、interrupt 屏障，零生产依赖，可嵌进任何宿主。

ma-nucho-pro / wife

Local-first, evidence-aware memory and project continuity for AI coding agents. Checkpoints, context packs and auditable receipts without telemetry.

BAJWA127 / torch-forge-trainer

Lightning-Fast PyTorch Training Toolkit 2026: Extend, Scale, Streamline

extracurricular-ai / codex-rewind

Unofficial distribution of OpenAI Codex CLI that adds /rewind and /redo — restores your files, not just the conversation. Git-free snapshots that never touch your git state. Tracks upstream releases; installs as `codexr` alongside the official build.

larr-bot / Townfall-Archive-Vault

Silent Hill Townfall Save Guardian 2026: Chapter Backup and Checkpoint Manager for Windows

zhuxun132 / InJepa

InJepa: Intention First Latent State Planning with Coupled JEPAs for Visual Navigation. Code, Clean150 tasks and E12 checkpoint.

DiverObserve / Stable-Diffusion-WebUI-Creative-Suite-Pro-Pack-Ultimate

Stable Diffusion WebUI Creative Suite Pro Pack Ultimate - Windows-focused build with a clean first-run path and complete professional modules enabled.

extracurricular-ai / dsh-filesnap

DSH Rewind & Redo — by FileSnap. 把对话和它改过的文件一起恢复到某一轮之前（受claude code rewind启发并改进）,不需要 git 仓库,并且可以撤销恢复（redo受opencode启发并改进），由rust驱动是市面上最快最可靠的dsh rewind插件. A blazing-fast rewind and redo plugin for DeepSeek Harness, powered by a 🦀 Rust core, tracking the conversation and the files it changed, no git required, low disk consumption

sudoingX / dgx-spark-deepseek-v41

DeepSeek-V4.1-Flash on two NVIDIA DGX Sparks: EXL3 weights from the official checkpoint, upstream vLLM with our patches and decode kernels

Greninja9257 / LabLLM

A native macOS lab for teaching tiny language models to think — build the architecture, train the weights, and watch a small LLM emerge from scratch, locally on Apple Silicon with custom data, tokenizers, checkpoints, and MLX acceleration.

AbrahamPaulJ / npuforge

Convert a Stable Diffusion 1.5 or SDXL checkpoint into a Qualcomm NPU model on the phone itself, with no PC. Supports LoRA merging, swapping LoRA/Controlnet/IPAdapters at runtime..

AmrLotfy / laravel-chores

Batched, checkpointed, resumable data operations for Laravel — backfills, cleanups, and migrations that survive deploys, crashes, and Ctrl+C.

JordyThery / Checkpoint

Checkpoint is a macOS app for Mac admins that cross-references devices between Apple Business or Apple School Manager and Jamf Pro, Jamf School or Intune.

MiaAI-Lab / Ornith-1.5-35B-A3B-DGX-Spark

Serve Ornith-1.5-35B-A3B (NVFP4 + in-checkpoint MTP, 256k ctx) on one DGX Spark — vLLM b12x recipe plus CUDA-graph fixes.

PerryLink / dsh-checkpoint-rewind

Claude Code /rewind for DeepSeek Harness — git-first workspace snapshots before every mutation, turn-boundary session forks, one-shot /rewind restore. A dsh-plugin capability seam.

Xuepoo / carryctx

Offline-first project context and durable team management for coding agents. A local SQLite CLI for tasks, sessions, checkpoints, worktrees, and read-only team context.

agent-substrate / always-on-agent

Run always-on messaging agents on Agent Substrate. A multi-tenant gateway holds the channel connections while each conversation runs as a suspendable actor, so presence stays warm and cognition suspends to reclaim compute. Reference implementation built on OpenClaw.