Search Results for "checkpoint"
Sort By:
Showing 85 open source projects for "checkpoint"
View related business solutions-
Veeam Data Platform v13.1 - Get Your Free TrialTry Veeam Data Platform today. Experience the unified platform that's secure by design, portable by default, and proven to recover clean, fast, and anywhere.
-
$300 Free Credits to Build on Google CloudPut your $300 in credit toward real workloads, then keep building with free monthly usage for 20+ products. No commitment and no charge until you upgrade.
-
1
Halo
Framework built by White Circle for training large language
...Performance features include FlashAttention, DeepEP, Liger kernels, Grouped GEMM, BF16 optimizers, LoRA, and QLoRA. RL rollouts can run through vLLM or SGLang while overlapping with training. Halo also provides Docker images, YAML recipes, checkpoint tools, dataset utilities, benchmarks, and agent-oriented technical documentation. -
2
VGGT-Ω
[CVPR 2026 Oral] VGGT Omega
...The project is associated with 3D understanding workflows where models infer scene geometry without a traditional multi-stage reconstruction pipeline. It includes pretrained model variants with different resolutions and text-alignment capabilities, though checkpoint access may require approval. The repository also provides a Gradio demo that can visualize predicted cameras and depth-unprojected point clouds as a GLB scene. VGGT-Omega is best suited for researchers and developers working on 3D reconstruction, visual geometry, and image-based scene understanding. -
3
Simple StyleGan2 for Pytorch
Simplest working implementation of Stylegan2
...By default, if the training gets cut off, it will automatically resume from the last checkpointed file. Once you have finished training, you can generate images from your latest checkpoint. If a previous checkpoint contained a better generator, (which often happens as generators start degrading towards the end of training), you can load from a previous checkpoint with another flag. A technique used in both StyleGAN and BigGAN is truncating the latent values so that their values fall close to the mean. The small the truncation value, the better the samples will appear at the cost of sample variety. -
4
Laya
Non-autoregressive System 1 decision engine
...It supports choice, score, and yes-or-no probability decisions in a single forward pass. The system works across more than 100 languages and uses a router to select an appropriate checkpoint for each request. It can run locally on CPU or GPU hardware and includes Python and TypeScript tooling. Integrations are available for HTTP, MCP, LangChain, LangGraph, LlamaIndex, CrewAI, and ONNX workflows. Laya is intended for fast classification, routing, triage, scoring, and other structured decisions where predictable outputs are more useful than generated prose. -
PRTG Catches Network Issues Before They Cause DowntimeReactive troubleshooting usually means hearing about a problem from frustrated users, not your monitoring tool. PRTG sets threshold-based alerts across devices, servers and applications, notifying your team by email, SMS or push the moment a metric crosses a set limit. That means catching a failing disk or overloaded server before it becomes an outage and getting time back from firefighting. Start a free trial and set your first alerts today.
-
5
HRM-Text
1B text generation model based on the HRM architecture
...The system combines hierarchical recurrent design, task-completion strengthening, and latent-space reasoning. Its training stack includes PrefixLM sequence packing, FlashAttention 3 kernels, PyTorch FSDP2, evaluation scripts, and checkpoint conversion tools. The repository supports reference pretraining runs for smaller and larger configurations, with Hopper-class GPUs expected for the attention path. It is useful for researchers and engineers exploring efficient language model pretraining, reasoning-focused architectures, and reproducible foundation model experiments. -
6
neural-style in TensorFlow
Neural style in TensorFlow
...The project uses TensorFlow automatic differentiation and the Adam optimizer rather than the original L-BFGS approach. Users can run it from the command line by providing a content image, style image inputs, and an output path. It supports checkpoint outputs, iteration control, style blending, and hyperparameter tuning for content weight, style weight, and learning rate. Overall, it is a focused research-style image generation tool for experimenting with artistic transfer and visual optimization. -
7
Megatron-LM
Ongoing research training transformer models at scale
...The framework includes mixed-precision training options such as FP16, BF16, FP8, and FP4 to maximize performance and memory efficiency on modern hardware. Megatron-LM is widely used in research and industry for pretraining GPT-, BERT-, T5-, and multimodal-style models, with tooling for checkpoint conversion and interoperability with Hugging Face. Overall, it is a production-grade system for organizations pushing the limits of large-scale language model training. -
8
PKHeX
Pokémon Save File Editor
...Pokémon Showdown sets and QR codes can be imported/exported to assist in sharing. PKHeX expects to save files that are not encrypted with console-specific keys. Use a save data manager to import and export save data from the console (Checkpoint, save_manager, JKSM, or SaveDataFiler). We do not support or condone cheating at the expense of others. Do not use significantly hacked Pokémon in battle or in trades with those who are unaware hacked Pokémon are in use. PKHeX is a Windows Forms application that requires .NET Framework v4.6, with experimental support for .NET 6.0. ... -
9
Transparent Background
This is a background removing tool powered by InSPyReNet
...We basically follow the virtual camera settings from pyvirtualcam. If you do not choose to install virtual camera, it will visualize real-time output with cv2.imshow. Use another checkpoint file. Default is trained with composite dataset and will be automatically downloaded if not available. -
Custom VMs From 1 to 96 vCPUs With 99.95% UptimeLive migration and automatic failover keep workloads online through maintenance. One free e2-micro VM every month.
-
10
AUTOMATIC1111 Stable Diffusion web UI
Stable Diffusion web UI
AUTOMATIC1111's stable-diffusion-webui is a powerful, user-friendly web interface built on the Gradio library that allows users to easily interact with Stable Diffusion models for AI-powered image generation. Supporting both text-to-image (txt2img) and image-to-image (img2img) generation, this open-source UI offers a rich feature set including inpainting, outpainting, attention control, and multiple advanced upscaling options. With a flexible installation process across Windows, Linux, and... -
11
Kimi K3 in C
A 2.78-trillion-parameter Kimi K3 running inference on a single CPU
Kimi K3 in C is a portable C99 inference engine built to run the 2.78-trillion-parameter Kimi K3 model on CPUs without BLAS, machine-learning frameworks, or GPUs. It demonstrates inference from a roughly 1.56 TB checkpoint with measured memory use as low as 8.24 GB. The runtime streams model trunk layers and routed experts from disk instead of keeping all weights resident in memory. Memory presets balance pinned layers and an expert LRU cache for laptops, desktops, workstations, and servers. Incremental generation preserves KV cache and recurrent state between tokens. ... -
12
TurboFieldfare
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
...Quantized weights, custom Metal kernels, chunked prefill, and a bounded expert cache improve efficiency. The installer downloads and repacks model ranges without staging a second full checkpoint. Its current scope is text-only Gemma inference on macOS 26 and Apple Silicon hardware. -
13
jlens
Companion code for the global workspace interpretability paper
Jacobian Lens is Anthropic’s reference implementation for examining what a language model’s internal activations are inclined to produce as text. It transports residual-stream vectors from selected layers and positions into the final-layer basis using an averaged input-output Jacobian. The transformed vectors are decoded through the model’s own unembedding into ranked vocabulary predictions. The package can fit new lenses, load saved ones, apply them to prompts, and merge results from... -
14
h3-metal
MiniMax H3 inference engine for Mac computers
...Users can trade speed, quality, and memory through denoising steps, layer counts, reuse modes, token reduction, internal render size, and SSD streaming. Optional terminal previews show intermediate and final frames during generation. The engine uses the original BF16 checkpoint and provides profiling tools for timing, Metal execution, tensor memory, allocations, and dispatch counts. -
15
Gemma in PyTorch
The official PyTorch implementation of Google's Gemma models
gemma_pytorch provides the official PyTorch reference for running and fine-tuning Google’s Gemma family of open models. It includes model definitions, configuration files, and loading utilities for multiple parameter scales, enabling quick evaluation and downstream adaptation. The repository demonstrates text generation pipelines, tokenizer setup, quantization paths, and adapters for low-rank or parameter-efficient fine-tuning. Example notebooks walk through instruction tuning and evaluation... -
16
Llama Recipes
Scripts for fine-tuning Meta Llama3 with composable FSDP & PEFT method
The 'llama-recipes' repository is a companion to the Meta Llama models. We support the latest version, Llama 3.1, in this repository. The goal is to provide a scalable library for fine-tuning Meta Llama models, along with some example scripts and notebooks to quickly get started with using the models in a variety of use-cases, including fine-tuning for domain adaptation and building LLM-based applications with Llama and other tools in the LLM ecosystem. The examples here showcase how to run... -
17
Harness-1
Ultra Recipe for Training Long-Horizon Search Agents
...This lets the policy focus on higher-level decisions instead of trying to keep every detail inside the model context. The repository includes inference utilities, training scripts, evaluation runners, dataset tools, and documentation for running the released checkpoint. Its main value is showing how a smaller open model can approach advanced search-agent behavior through structured retrieval state and reinforcement learning. -
18
Laya-MLX
Native MLX runtime for Laya typed decision models
...Its implementation moves the encoder, decision transformer, scoring head, and action head into MLX. The project reports short-decision latency in the single-digit to low-teens millisecond range on an M3 Max, depending on the checkpoint. It also includes routing utilities, demos, validation tests, benchmarks, and a Python API. -
19
AI Dev Tasks
A simple task management system for managing AI dev agents
AI Dev Tasks is a Markdown-based task management system for AI-assisted software development. It provides structured prompt files that guide an AI coding assistant through feature planning, PRD creation, task breakdown, and implementation checkpoints. The project is designed to work with AI-powered IDEs and CLI agents such as Amp, Claude Code, Windsurf, and similar tools. Instead of asking an agent to jump directly into code, users can follow a staged process that clarifies requirements... -
20
opcode
A powerful GUI app and Toolkit for Claude Code
opcode is an open source desktop application and toolkit designed to enhance the developer experience when working with Claude Code by providing a graphical interface and advanced workflow management tools. The project acts as a command center for AI-assisted programming, bridging the gap between command-line workflows and modern visual development environments. Built using the Tauri framework, Opcode enables developers to manage multiple Claude sessions, create custom agents, and track... -
21
Coconut
Training Large Language Model to Reason in a Continuous Latent Space
Coconut is the official PyTorch implementation of the research paper “Training Large Language Models to Reason in a Continuous Latent Space.” The framework introduces a novel method for enhancing large language models (LLMs) with continuous latent reasoning steps, enabling them to generate and refine reasoning chains within a learned latent space rather than relying solely on discrete symbolic reasoning. It supports training across multiple reasoning paradigms—including standard... -
22
Scrapling
An adaptive Web Scraping framework
Scrapling is an adaptive web scraping framework designed to handle everything from a single HTTP request to large-scale, concurrent crawls. Built for modern websites, it intelligently adapts to structural changes by automatically relocating elements when page layouts update. The framework includes advanced fetchers capable of bypassing anti-bot protections such as Cloudflare Turnstile using stealth and browser automation techniques. Its powerful spider system supports multi-session crawling,... -
23
Zypher Agent
A minimal yet powerful framework for creating AI agents
Zypher Agent is an open-source framework for building full-featured AI agents that can be embedded directly into applications, enabling reactive decision loops where the agent dynamically chooses its next actions. Unlike workflow-style orchestrators, it uses a reactive agent loop that interprets the task, reasons about next steps via LLMs, and integrates directly with extensible tools and external services. Zypher prioritizes native support for multiple model providers such as OpenAI and... -
24
fvcore
Collection of common code shared among different research projects
fvcore is a lightweight utility library that factors out common performance-minded components used across Facebook/Meta computer-vision codebases. It provides numerics and loss layers (e.g., focal loss, smooth-L1, IoU/GIoU) implemented for speed and clarity, along with initialization helpers and normalization layers for building PyTorch models. Its common modules include timers, logging, checkpoints, registry patterns, and configuration helpers that reduce boilerplate in research code. A... -
25
seaart-save-metadata
Chrome extention to save metadata of works on seart.ai in XMP in files
...Chrome extention to save metadata of works on seart.ai in XMP and embed them into files. Works on /explore/detail pages and /postDetail/ pages. Does NOT work to videos. Filenames are automatically sanitized. It saves: prompt, model, checkpoint, lora, creator, source, etc. of each work. Press a button to start fetching metadata and download a work on a target page. # Installation Unpack the release and directly load it into Chrome with "Load unpacked".
