Search Results for "weight"
Sort By:
Showing 38 open source projects for "weight"
View related business solutions-
Veeam Data Platform v13.1Try Veeam Data Platform today. Experience the unified platform that's secure by design, portable by default, and proven to recover clean, fast, and anywhere.
-
99.99% Uptime for MySQL and PostgreSQL DatabasesCloud SQL Enterprise Plus delivers near-zero downtime with 35 days of point-in-time recovery. Supports MySQL, PostgreSQL, and SQL Server.
-
1
Reef
Continual learning infra for self-improving agents
...The system can improve either underlying model weights or the surrounding agent harness, including prompts, skills, and rules. Its core loop follows four stages: serve requests, observe feedback, grow candidate improvements, and commit accepted updates. Weight-training workflows can integrate with systems such as Slime and SGLang, while harness optimization can run without local training GPUs. Versioned artifacts let deployments stay operational while new candidates are evaluated and released. Reef also supports test-time learning scenarios where measurable task outcomes can drive iterative improvement. -
2
Gemma
Gemma open-weight LLM library, from Google DeepMind
...Through included tutorials and Colab notebooks, users can explore examples covering sampling, multi-modal interactions, and fine-tuning workflows. By providing accessible open-weight models, Gemma enables researchers and developers to experiment with state-of-the-art LLM architectures. -
3
FLUX.2
Official inference repo for FLUX.2 models
FLUX.2 is a state-of-the-art open-weight image generation and editing model released by Black Forest Labs aimed at bridging the gap between research-grade capabilities and production-ready workflows. The model offers both text-to-image generation and powerful image editing, including editing of multiple reference images, with fidelity, consistency, and realism that push the limits of what open-source generative models have achieved. -
4
OpenVoice
Instant voice cloning by MIT and MyShell. Audio foundation model
...Architecturally, OpenVoice separates “tone color” cloning from style control, which makes it easier to keep a consistent identity while flexibly changing prosody or language. The project provides open-weight models, inference code, and examples, making it suitable both for research and for building production voice experiences. It is actively developed by MyShell, which also integrates OpenVoice into broader agent and entertainment workflows. -
Train ML Models With SQL You Already KnowBuild and deploy ML models using familiar SQL. Automate data prep with built-in Gemini. Query 1 TB and store 10 GB free monthly.
-
5
gpt-oss
gpt-oss-120b and gpt-oss-20b are two open-weight language models
gpt-oss is OpenAI’s open-weight family of large language models designed for powerful reasoning, agentic workflows, and versatile developer use cases. The series includes two main models: gpt-oss-120b, a 117-billion parameter model optimized for general-purpose, high-reasoning tasks that can run on a single H100 GPU, and gpt-oss-20b, a lighter 21-billion parameter model ideal for low-latency or specialized applications on smaller hardware. -
6
MiniMax-M1
Open-weight, large-scale hybrid-attention reasoning model
MiniMax-M1 is presented as the world’s first open-weight, large-scale hybrid-attention reasoning model, designed to push the frontier of long-context, tool-using, and deeply “thinking” language models. It is built on the MiniMax-Text-01 foundation and keeps the same massive parameter budget, but reworks the attention and training setup for better reasoning and test-time compute scaling. -
7
FLUX.1
Official inference repo for FLUX.1 models
FLUX.1 repository contains inference code and tooling for the FLUX.1 text-to-image diffusion models, enabling developers and researchers to generate and edit images from natural-language prompts using open-weight versions of the model on their own hardware or within custom applications. The project is part of a larger family of FLUX models developed by Black Forest Labs, designed to produce high-quality, detailed visuals from text descriptions with competitive prompt adherence and artistic fidelity. This repo focuses on running the open-source model variants efficiently, providing scripts, model loading logic, and examples for local installations, and supports integration with Python toolchains like PyTorch and popular generative pipelines. ... -
8
Ideogram 4
Open image model at the forefront of design
Ideogram 4 is an open-weight text-to-image model focused on high-quality visual generation, design control, and accurate text rendering inside images. It is built for users who need more than generic image generation, especially when layout, typography, composition, color, and language understanding matter. The project introduces a structured JSON prompting workflow that gives creators more explicit control over scene details and visual constraints. -
9
AIMET
AIMET is a library that provides advanced quantization and compression
Qualcomm Innovation Center (QuIC) is at the forefront of enabling low-power inference at the edge through its pioneering model-efficiency research. QuIC has a mission to help migrate the ecosystem toward fixed-point inference. With this goal, QuIC presents the AI Model Efficiency Toolkit (AIMET) - a library that provides advanced quantization and compression techniques for trained neural network models. AIMET enables neural networks to run more efficiently on fixed-point AI hardware... -
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
10
CogVideo
Text and image to video generation: CogVideoX and CogVideo
CogVideo is an open-source family of advanced video generation models that can create videos from text, images, or existing video inputs. Built on large-scale Transformer and diffusion architectures, it enables multimodal generation across text-to-video, image-to-video, and video continuation tasks. The latest CogVideoX models offer higher resolution outputs, longer video durations, and improved controllability through prompt engineering. The project includes tools for inference,... -
11
Mistral Inference
Official inference library for Mistral models
Open and portable generative AI for devs and businesses. We release open-weight models for everyone to customize and deploy where they want it. Our super-efficient model Mistral Nemo is available under Apache 2.0, while Mistral Large 2 is available through both a free non-commercial license, and a commercial license. -
12
mac code
Claude Code, but it runs on your Mac for free
mac code is a local AI coding agent designed to run large language models directly on Apple Silicon machines without relying on cloud services, effectively transforming a Mac into a self-contained AI development environment. The project focuses on enabling models that traditionally exceed available RAM to run efficiently by streaming model weights from SSD storage, thereby overcoming hardware limitations through innovative memory management techniques. It operates as a CLI-based assistant... -
13
OpenAI Harmony
Renderer for the harmony response format to be used with gpt-oss
...For users accessing gpt-oss through third-party providers like HuggingFace, Ollama, or vLLM, Harmony formatting is handled automatically, but developers building custom inference setups must implement it directly. With its flexible design, Harmony serves as the foundation for creating more interpretable, controlled, and extensible interactions with open-weight language models. -
14
Deep Agents
The batteries-included agent harness
...It provides an opinionated agent setup while allowing developers to override or replace individual pieces without forking the project. The framework is model-agnostic and works with tool-calling models from hosted providers, open-weight deployments, or local runtimes. Built on LangGraph, it includes persistence, checkpointing, streaming, and production-oriented orchestration. Agents can delegate to sub-agents, work with files, run shell commands, manage long contexts, and retain memory across sessions. It also supports human approval of tool calls, reusable skills, custom tools, and MCP servers. -
15
OpenWorker
AI that gets your everyday tasks done
...The agent breaks requested outcomes into steps and works across local files, the terminal, desktop resources, and connected applications. Users can bring API keys for multiple commercial or open-weight model providers, or run locally through Ollama. More than 25 integrations connect services such as GitHub, Slack, Jira, Notion, Gmail, Outlook, and Google Calendar. OpenWorker requests approval before consequential actions such as sending messages, changing events, or running commands. Scheduled automations and Slack-triggered sessions extend it to recurring and team-based workflows. -
16
TabFM
scikit-learn compatible tabular foundation model
TabFM is a tabular foundation model from Google Research for zero-shot classification and regression on structured datasets. It is designed to work with mixed numerical and categorical columns without requiring a custom training run for every new table. Instead of fitting model weights to the user’s dataset, TabFM uses in-context learning by reading training examples and test rows together at inference time. The library provides scikit-learn-compatible classifier and regressor interfaces,... -
17
OpenAI Privacy Filter
Bidirectional token-classification model for identifiable info
OpenAI Privacy Filter is an open-weight machine learning model designed to detect and mask personally identifiable information in text with high efficiency and contextual awareness. It operates as a bidirectional token classification system that labels sensitive data in a single forward pass rather than generating text sequentially, enabling fast processing for large datasets. -
18
MatMul-Free LM
Implementation for MatMul-free LM
MatMul-Free LM is an experimental implementation of a large language model architecture designed to eliminate traditional matrix multiplication operations used in transformer networks. Since matrix multiplication is one of the most computationally expensive components of modern language models, the project explores alternative computational strategies that reduce hardware requirements while maintaining comparable performance. The architecture relies on quantization-aware training and... -
19
LLM-Pruner
On the Structural Pruning of Large Language Models
...The framework relies on gradient-based analysis to determine which parameters contribute least to model performance, enabling targeted structural pruning rather than simple weight removal. After pruning, the framework applies lightweight fine-tuning methods such as LoRA to recover performance using relatively small datasets and short training times. -
20
SMILI
Scientific Visualisation Made Easy
The Simple Medical Imaging Library Interface (SMILI), pronounced 'smilie', is an open-source, light-weight and easy-to-use medical imaging viewer and library for all major operating systems. The main sMILX application features for viewing n-D images, vector images, DICOMs, anonymizing, shape analysis and models/surfaces with easy drag and drop functions. It also features a number of standard processing algorithms for smoothing, thresholding, masking etc. images and models, both with graphical user interfaces and/or via the command-line. ... -
21
Bootstrap Your Own Latent (BYOL)
Usable Implementation of "Bootstrap Your Own Latent" self-supervised
...This repository offers a module that one can easily wrap any image-based neural network (residual network, discriminator, policy network) to immediately start benefitting from unlabelled image data. There is now new evidence that batch normalization is key to making this technique work well. A new paper has successfully replaced batch norm with group norm + weight standardization, refuting that batch statistics are needed for BYOL to work. Simply plugin your neural network, specifying (1) the image dimensions as well as (2) the name (or index) of the hidden layer, whose output is used as the latent representation used for self-supervised training. -
22
BitNet
BitNet: Scaling 1-bit Transformers for Large Language Models
...The architecture introduces specialized layers such as BitLinear, which replace standard linear projections in transformer networks with quantized operations. By limiting weight precision while maintaining efficient scaling and normalization strategies, the architecture aims to retain competitive performance while significantly reducing hardware requirements. -
23
llama3-from-scratch-zh
Implementing an llama3 Chinese version from scratch
Llama3 from Scratch zh is an open-source educational implementation that explains the internal architecture of Meta's Llama 3 language model through step-by-step Python and PyTorch code. It reconstructs major Transformer components using explicit tensor operations and matrix calculations. The project demonstrates how to load pretrained Llama 3 weights and examine their structure. Readers can explore tokenization, embedding layers, attention calculations, normalization, and feed-forward... -
24
DeepSeek VL
Towards Real-World Vision-Language Understanding
DeepSeek-VL is DeepSeek’s initial vision-language model that anchors their multimodal stack. It enables understanding and generation across visual and textual modalities—meaning it can process an image + a prompt, answer questions about images, caption, classify, or reason about visuals in context. The model is likely used internally as the visual encoder backbone for agent use cases, to ground perception in downstream tasks (e.g. answering questions about a screenshot). The repository... -
25
FastChat
Open platform for training, serving, and evaluating language models
FastChat is an open platform for training, serving, and evaluating large language model-based chatbots. If you do not have enough memory, you can enable 8-bit compression by adding --load-8bit to the commands above. This can reduce memory usage by around half with slightly degraded model quality. It is compatible with the CPU, GPU, and Metal backend. Vicuna-13B with 8-bit compression can run on a single NVIDIA 3090/4080/T4/V100(16GB) GPU. In addition to that, you can add --cpu-offloading to...
