Search Results for "so" - Page 17
Sort By:
Showing 1303 open source projects for "so"
View related business solutions-
$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.
-
Fully Managed MySQL, PostgreSQL, and SQL ServerCloud SQL handles your database ops end to end, so you can focus on your app.
-
1
Atheris
A Coverage-Guided, Native Python Fuzzer
...Because many Python libraries are thin wrappers over C/C++ code, Atheris is equally adept at surfacing memory safety issues in extension modules compiled with sanitizers. The tool integrates smoothly with Python’s packaging and unit-test ecosystems, so you can wrap existing tests as fuzz targets and keep results understandable. It supports structured input strategies and custom mutators, which is especially helpful for text and data formats common in Python workloads. In practice, Atheris compresses weeks of edge-case brainstorming into hours of automated exploration with actionable, minimized reproductions. -
2
Tunix
A JAX-native LLM Post-Training Library
Tunix is a JAX-native library for post-training large language models, bringing supervised fine-tuning, reinforcement learning–based alignment, and knowledge distillation into one coherent toolkit. It embraces JAX’s strengths—functional programming, jit compilation, and effortless multi-device execution—so experiments scale from a single GPU to pods of TPUs with minimal code changes. The library is organized around modular pipelines for data loading, rollout, optimization, and evaluation, letting practitioners swap components without rewriting the whole stack. Examples and reference configs demonstrate end-to-end runs for common model families, helping teams reproduce baselines before customizing. ... -
3
latexify
A library to generate LaTeX expression from Python code
...The tool aims to preserve semantics such as exponentiation, summations, products, piecewise definitions, and function application while hiding Pythonic scaffolding. Users can control rendering details for names and operators so the output conforms to a project’s notation style. Because it works directly from Python source, it enables a tight authoring loop. -
4
4M
4M: Massively Multimodal Masked Modeling
...Training/inference configs and issues discuss things like depth tokenizers, input masks for generation, and CUDA build questions, signaling active research iteration. The design leans into flexibility and steerability, so prompts and masks can shape behavior without bespoke heads per task. In short, 4M provides a unified recipe to pretrain large multimodal models that generalize broadly while remaining practical to fine-tune. -
Host LLMs in Production With On-Demand GPUsDeploy your model, get an endpoint, pay only for compute time. No GPU provisioning or infrastructure management required.
-
5
ML Ferret
Refer and Ground Anything Anywhere at Any Granularity
Ferret is Apple’s end-to-end multimodal large language model designed specifically for flexible referring and grounding: it can understand references of any granularity (boxes, points, free-form regions) and then ground open-vocabulary descriptions back onto the image. The core idea is a hybrid region representation that mixes discrete coordinates with continuous visual features, so the model can fluidly handle “any-form” referring while maintaining precise spatial localization. The repo presents the vision-language pipeline, model assets, and paper resources that show how Ferret answers questions, follows instructions, and returns grounded outputs rather than just text. In practice, this enables tasks like “find that small red icon next to the chart and describe it” where both the linguistic reference and the visual region are ambiguous without fine spatial reasoning. -
6
Synthetic Data Kit
Tool for generating high quality Synthetic datasets
...The kit’s design goal is to shorten the “data prep” bottleneck by turning dataset creation into a repeatable pipeline rather than ad-hoc notebooks. It supports generation of rationales/chain-of-thought variants, configurable sampling, and guardrails so outputs meet format constraints and quality checks. Examples and guides show how to target task-specific behaviors like tool use or step-by-step reasoning, then save directly into training-ready files. -
7
Multimodal
TorchMultimodal is a PyTorch library
...The design emphasizes composability: you can mix and match encoder, fusion, and decoder components rather than starting from monolithic models. The repository also includes example scripts and datasets for common multimodal tasks (e.g. retrieval, visual question answering, grounding) so you can test and compare models end to end. Installation supports both CPU and CUDA, and the codebase is versioned, tested, and maintained. -
8
ImageBind
ImageBind One Embedding Space to Bind Them All
ImageBind is a multimodal embedding framework that learns a shared representation space across six modalities—images, text, audio, depth, thermal, and IMU (inertial motion) data—without requiring explicit pairwise training for every modality combination. Instead of aligning each pair independently, ImageBind uses image data as the central binding modality, aligning all other modalities to it so they can interoperate zero-shot. This creates a unified embedding space where representations from any modality can be compared or retrieved against any other (e.g., matching sound to text or depth to image). The model is trained using large-scale contrastive learning, leveraging diverse datasets from natural images, videos, audio clips, and sensor data. ... -
9
GenAI Processors
GenAI Processors is a lightweight Python library
...Processors can be composed sequentially (to build multi-step flows) or in parallel (to fan-out work and merge results), which makes sophisticated agent behaviors easy to express with simple operators. The library offers built-in processors for classic turn-based Gemini calls as well as Live API streaming, so you can mix “batch” and real-time interactions in the same graph. It leans on Python’s asyncio to coordinate concurrency, handle network I/O, and juggle background compute threads without blocking. -
Earn up to 16% annual interest with Nexo.Generate interest, borrow against your crypto, and trade a range of cryptocurrencies — all in one platform. Geographic restrictions, eligibility, and terms apply.
-
10
PydanticAI
Agent Framework / shim to use Pydantic with LLMs
When I first found FastAPI, I got it immediately. I was excited to find something so innovative and ergonomic built on Pydantic. Virtually every Agent Framework and LLM library in Python uses Pydantic, but when we began to use LLMs in Pydantic Logfire, I couldn't find anything that gave me the same feeling. PydanticAI is a Python Agent Framework designed to make it less painful to build production-grade applications with Generative AI. -
11
Kubeflow pipelines
Machine Learning Pipelines for Kubeflow
...A pipeline component is a self-contained set of user code, packaged as a Docker image, that performs one step in the pipeline. For example, a component can be responsible for data preprocessing, data transformation, model training, and so on. -
12
gusty
Making DAG construction easier
...All you have to do is provide a list of dependencies or external_dependencies inside of a task file, and gusty will automatically set each task's dependencies and create external task sensors for any external dependencies listed. gusty works with both Airflow 1.x and Airflow 2.x, and has even more features, all of which aim to make the creation, management, and iteration of DAGs more fluid, so that you can intuitively design your DAG and build your tasks. -
13
Gretel Synthetics
Synthetic data generators for structured and unstructured text
...Generate unlimited data in minutes with synthetic data delivered as-a-service. Synthesize data that are as good or better than your original dataset, and maintain relationships and statistical insights. Customize privacy settings so that data is always safe while remaining useful for downstream workflows. Ensure data accuracy and privacy confidently with expert-grade reports. Need to synthesize one or multiple data types? We have you covered. Even take advantage or multimodal data generation. Synthesize and transform multiple tables or entire relational databases. ... -
14
Offensive Web Testing Framework
Offensive Web Testing Framework (OWTF), is a framework
OWASP OWTF is a project focused on penetration testing efficiency and alignment of security tests to security standards like the OWASP Testing Guide (v3 and v4), the OWASP Top 10, PTES and NIST so that pentesters will have more time to see the big picture and think out of the box. More efficiently find, verify and combine vulnerabilities. Have time to investigate complex vulnerabilities like business logic/architectural flaws or virtual hosting sessions. Perform more tactical/targeted fuzzing on seemingly risky areas. Demonstrate true impact despite the short timeframes we are typically given to test. ... -
15
django-environ
Django-environ allows you to utilize 12factor inspired environment
The idea of this package is to unify a lot of packages that make the same stuff: Take a string from os.environ, parse and cast it to some of useful python typed variables. To do that and to use the 12factor approach, some connection strings are expressed as url, so this package can parse it and return a urllib.parse.ParseResult. These strings from os.environ are loaded from a .env file and filled in os.environ with setdefault method, to avoid overwriting the real environment. A similar approach is used in Two Scoops of Django book and explained in the 12factor-Django article. django-environ is the Python package that allows you to use the Twelve-factor methodology to configure your Django application with environment variables.For that, it gives you an easy way to configure Django application using environment variables obtained from an environment file and provided by the OS. -
16
Deepchecks
Test Suites for validating ML models & data
Deepchecks is the leading tool for testing and for validating your machine learning models and data, and it enables doing so with minimal effort. Deepchecks accompany you through various validation and testing needs such as verifying your data’s integrity, inspecting its distributions, validating data splits, evaluating your model and comparing between different models. While you’re in the research phase, and want to validate your data, find potential methodological problems, and/or validate your model and evaluate it. ... -
17
PML
The easiest way to use deep metric learning in your application
...In the diagram below, a miner finds the indices of hard pairs within a batch. These are used to index into the distance matrix, computed by the distance object. For this diagram, the loss function is pair-based, so it computes a loss per pair. -
18
bidict
The bidirectional mapping library for Python
...One of the best things about bidict is that it touches a surprising number of interesting Python corners, especially given its small size and scope. Choose a tier and GitHub handles everything else. Your GitHub sponsorship will automatically go on the same bill you already have set up with GitHub, so after the one-click signup, there’s nothing else to do. -
19
Selenium-python Helium
Selenium-python but lighter: Helium is the best Python library
...What's more, they are easier to read and more stable with respect to changes in the underlying web page. Selenium-python is great for web automation. Helium makes it easier to use. Helium ships with its own copies of ChromeDriver and geckodriver so you don't need to download and put them on your PATH. Unlike Selenium, Helium lets you interact with elements inside nested iFrames, without having to first "switch to" the iFrame. Helium notices when popups open or close and focuses / defocuses them like a user would. You can also easily switch to a window by (parts of) its title. -
20
TensorFlow Model Garden
Models and examples built with TensorFlow
The TensorFlow Model Garden is a repository with a number of different implementations of state-of-the-art (SOTA) models and modeling solutions for TensorFlow users. We aim to demonstrate the best practices for modeling so that TensorFlow users can take full advantage of TensorFlow for their research and product development. To improve the transparency and reproducibility of our models, training logs on TensorBoard.dev are also provided for models to the extent possible though not all models are suitable. A flexible and lightweight library that users can easily use or fork when writing customized training loop code in TensorFlow 2.x. ... -
21
Eel
A Python library for making simple Electron-like HTML/JS GUI apps
Eel is a little Python library for making simple Electron-like offline HTML/JS GUI apps, with full access to Python capabilities and libraries. Eel hosts a local webserver, then lets you annotate functions in Python so that they can be called from Javascript, and vice versa. Eel is designed to take the hassle out of writing short and simple GUI applications. If you are familiar with Python and web development, probably just jump to this example which picks random file names out of the given folder (something that is impossible from a browser). There are several options for making GUI apps in Python, but if you want to use HTML/JS (in order to use jQueryUI or Bootstrap, for example) then you generally have to write a lot of boilerplate code to communicate from the Client (Javascript) side to the Server (Python) side. -
22
Local File Organizer
An AI-powered file management tool that ensures privacy
...The project focuses on privacy-first file organization by performing all processing locally rather than sending data to external cloud services. It uses language and vision models to understand the contents of documents, images, and other file types so that files can be grouped intelligently according to their meaning or context. The system scans directories, extracts relevant information from files, and restructures folder hierarchies to make content easier to locate and manage. Through AI-driven analysis, the software can detect themes, topics, and metadata in files, allowing it to organize information in ways that traditional rule-based file managers cannot achieve. ... -
23
PyScaffold
Python project template generator with batteries included
...After cd-ing into your new project and creating (or activating) an isolated development environment (with virtualenv, conda or your preferred tool), you can do the usual editable install. All configuration can be done in setup.cfg like changing the description, URL, classifiers, installation requirements and so on as defined by setuptools. That means in most cases it is not necessary to tamper with setup.py. -
24
douyin
Open source Douyin crawler for collecting and downloading public data
...It includes multiple usage modes such as a desktop GUI, a web service interface, and a command line tool for flexible deployment. In addition to data collection, it supports incremental updates so users can track and gather newly published content without reprocessing previously collected data. -
25
JamAI Base
The collaborative spreadsheet for AI
...The platform integrates both a relational database and a vector database into a single embedded architecture, allowing developers to store structured data alongside semantic embeddings. It includes built-in orchestration for large language models, vector search, and reranking pipelines so that AI applications can retrieve relevant information before generating responses. JamAI Base exposes its functionality through a simple REST API and a spreadsheet-style interface that allows users to manage AI workflows visually. One of the key ideas behind the platform is the concept of generative tables, which allow database columns to automatically populate with AI-generated content. ...
