Search Results for "context" - Page 7
Sort By:
Showing 459 open source projects for "context"
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.
-
Build Data Resilience - Take the Assessment TodayIs your recovery strategy as strong as you think? Take this quick self-assessment to check your recovery readiness and gain tailored insights. In only 2 minutes, you'll learn where you fall on the recovery readiness scale.
-
1
PrivateGPT
Interact with your documents using the power of GPT
PrivateGPT is a production-ready, privacy-first AI system that allows querying of uploaded documents using LLMs, operating completely offline in your own environment. It provides contextual generative AI capabilities without sending data externally. Now maintained under Zylon.ai with enterprise deployment options (air gapped, cloud, or on-prem). -
2
Hindsight
Hindsight: Agent Memory That Learns
...Instead of relying solely on vector similarity or basic retrieval techniques, Hindsight organizes information into distinct categories such as facts, experiences, beliefs, and observations, allowing agents to differentiate between raw data and inferred knowledge. The system operates through three core mechanisms—retain, recall, and reflect—which respectively handle storing information, retrieving relevant context, and generating new insights based on accumulated experience. -
3
GLM-4.5
GLM-4.5: Open-source LLM for intelligent agents by Z.ai
...GLM-4.5 achieves strong performance on 12 industry-standard benchmarks, ranking 3rd overall, while GLM-4.5-Air balances competitive results with greater efficiency. The models support FP8 and BF16 precision, and can handle very large context windows of up to 128K tokens. Flexible inference is supported through frameworks like vLLM and SGLang with tool-call and reasoning parsers included. -
4
SEO Machine
A specialized Claude Code workspace for creating long-form
...It incorporates real data sources like Google Analytics and Search Console to guide decision-making and improve content effectiveness. The architecture emphasizes context-awareness, using brand voice, style guides, and keyword strategies to maintain consistency across outputs. It also includes performance evaluation tools that score content and suggest improvements before publishing. -
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.
-
5
AI Agent Deep Dive
AI Agent Source Code Deep Research Report
...The project is organized as a learning resource rather than a standalone framework, making it particularly useful for developers who want to move beyond surface-level prompt engineering into full agent system design. It explores how agents interact with environments, execute tasks, and maintain context over time, highlighting both strengths and limitations of current approaches. The repository likely includes diagrams, annotated code samples, and conceptual walkthroughs that mirror real production systems. -
6
Youtu-Agent
A simple yet powerful agent framework that delivers with models
...The system focuses on reducing the complexity traditionally involved in configuring large language model agents by providing a modular architecture that separates execution environments, tools, and context management. This structure allows developers to rapidly assemble agent systems capable of performing tasks such as research, file processing, and data analysis. The framework supports automated generation of agent components, enabling the system to synthesize prompts, tool interfaces, and workflow configurations automatically. Youtu-Agent also incorporates hybrid learning strategies that combine experience accumulation with reinforcement learning to improve agent performance over time. ... -
7
StoryMem
Official code for StoryMem: Multi-shot Long Video Storytelling
StoryMem is a narrative-focused memory accumulation system that lets users build, store, and reference past conversational context or story elements with an AI, effectively enabling the AI to maintain and recall personalized story memories or character arcs over time. Instead of treating each interaction as stateless, it tracks user-defined memory nodes, tags, and story threads so that future interactions can draw on established narrative context like character traits, past events, or ongoing plotlines. ... -
8
UltraRAG
Less Code, Lower Barrier, Faster Deployment
UltraRAG 2.0 is a low-code, MCP-enabled RAG framework that aims to lower the barrier to building complex retrieval pipelines for research and production. It provides end-to-end recipes—from encoding and indexing corpora to deploying retrievers and LLMs—so users can reproduce baselines and iterate rapidly. The toolkit comes with built-in support for popular RAG datasets, large corpora, and canonical baselines, plus documentation that walks from “quick start” to debugging and case analysis. It... -
9
Django Rules
Awesome Django authorization, without the database
...No need to mess around with a database to figure out whether John really wrote that book. Simple. Dive in the code. You'll need 10 minutes to figure out how it works. Powerful. rules comes complete with advanced features, such as invocation context and storage for arbitrary data, skipping evaluation of predicates. -
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.
-
10
Learn Claude Code
Bash is all you need, write a claude code with only 16 line code
...The goal is to demystify agent architectures like Claude Code by having learners build simplified versions themselves and observe how tools, memory management, planning constraints, and context isolation contribute to reliable agent behavior. Along the way, the project teaches fundamentals such as how to let models call external tools, maintain clean memory for long tasks, and inject domain expertise without retraining the model. -
11
vJEPA-2
PyTorch code and models for VJEPA2 self-supervised learning from video
VJEPA2 is a next-generation self-supervised learning framework for video that extends the “predict in representation space” idea from i-JEPA to the temporal domain. Instead of reconstructing pixels, it predicts the missing high-level embeddings of masked space-time regions using a context encoder and a slowly updated target encoder. This objective encourages the model to learn semantics, motion, and long-range structure without the shortcuts that pixel-level losses can invite. The architecture is designed to scale: spatiotemporal ViT backbones, flexible masking schedules, and efficient sampling let it train on long clips while remaining stable. ... -
12
DeepSeek Coder
DeepSeek Coder: Let the Code Write Itself
...The models are trained from scratch on a massive corpus (~2 trillion tokens), of which about 87% is code and 13% is natural language. This dataset covers project-level code structure (not just line-by-line snippets), using a large context window (e.g. 16K) and a secondary fill-in-the-blank objective to encourage better contextual completions and infilling. Multiple sizes of the model are offered (e.g. 1B, 5.7B, 6.7B, 33B) so users can trade off inference cost vs capability. The repo provides model weights, documentation on training setup, evaluation results on common benchmarks (HumanEval, MultiPL-E, APPS, etc.), and inference tools. -
13
Megatron-LM
Ongoing research training transformer models at scale
...The repository provides both a reference training implementation and Megatron Core, a composable library of high-performance building blocks for custom large-model pipelines. It supports advanced parallelism strategies including tensor, pipeline, data, expert, and context parallelism, enabling training across massive multi-GPU and multi-node clusters. 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. ... -
14
book-to-skill
Turn any technical book PDF into a Claude Code skill
book-to-skill is a Claude Code skill that turns technical books and documents into reusable AI reference skills. It extracts content from PDFs and EPUBs, then organizes the material so an assistant can study, reference, and apply it while working. The project is useful for transforming dense manuals, textbooks, internal documentation, or technical guides into practical agent-accessible knowledge. It includes an extraction script and a SKILL.md workflow that guides how the resulting content... -
15
Open Notebook
An Open Source implementation of Notebook LM with more flexibility
...Open Notebook enables users to organize and analyze multi-modal content such as PDFs, videos, audio files, web pages, and Office documents. It combines full-text and vector search with context-aware AI chat to deliver insights grounded in your own research materials. With advanced features like multi-speaker podcast generation, customizable content transformations, and a comprehensive REST API, Open Notebook provides a powerful and extensible research environment. -
16
Aider
Aider is AI pair programming in your terminal
Aider is an AI pair programming tool that runs directly in your terminal, helping developers build new projects or extend existing codebases faster and more confidently. It works alongside you like a coding partner, using powerful large language models to understand your code and implement precise changes. Aider creates a structured map of your entire repository, allowing it to handle large and complex projects effectively. It supports over 100 programming languages, making it flexible for... -
17
Inter
The Inter font family
Inter is a typeface carefully crafted & designed for computer screens. Inter features a tall x-height to aid in readability of mixed-case and lower-case text. Several OpenType features are provided as well, like contextual alternates that adjusts punctuation depending on the shape of surrounding glyphs, slashed zero for when you need to disambiguate "0" from "o", tabular numbers, etc. Using Inter is as easy as downloading & installing the font files. There's of course no absolute right or... -
18
Qwen-Agent
Agent framework and applications built upon Qwen>=3.0
...It ships with example applications (Browser Assistant, Code Interpreter, Custom Assistant), supports GUI front-ends, backends, server setups. Agent workflow can maintain context / memory to perform multi-turn or more complex logic over time. It acts as the backend for Qwen Chat among other use cases. Built-in Code Interpreter tool that can execute code (locally) as part of agent workflows. -
19
OpenAI Privacy Filter
Bidirectional token-classification model for identifiable info
...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. The model supports long-context inputs, allowing it to analyze extensive documents without chunking, which improves consistency in redaction tasks. It can run locally on standard hardware, ensuring that sensitive information never leaves the user’s environment and supporting privacy-first workflows. The system is fine-tunable, enabling adaptation to specific datasets or compliance requirements across industries. ... -
20
MiniCPM4.1
Achieving 3+ generation speedup on reasoning tasks
...The model also supports both dense and sparse attention mechanisms, enabling more efficient computation depending on the selected inference framework. With improved pretraining on longer sequences and enhanced scaling techniques, MiniCPM4.1 delivers better performance in long-context tasks and complex problem solving. -
21
MiniCPM4
Ultra-Efficient LLMs on End Device
...Unlike traditional large-scale models that require extensive computational resources, MiniCPM4 focuses on delivering competitive reasoning and language capabilities while maintaining significantly lower latency and higher efficiency. It achieves this through optimized architectures, scalable training strategies, and techniques such as long-context pretraining and YaRN-based length extension, allowing it to handle sequences up to 128K tokens effectively. The model demonstrates strong performance across tasks such as long-text comprehension, reasoning, and general language generation, often outperforming similar-sized models in both speed and accuracy. MiniCPM4 is available in multiple parameter sizes, making it adaptable to different deployment scenarios ranging from mobile devices to GPUs. -
22
Mito
AI-powered Jupyter spreadsheet that converts workflows into Python
Mito is an open source set of Jupyter extensions designed to speed up Python workflows and data analysis. It combines a spreadsheet-style interface with AI-assisted coding, allowing users to explore, clean, and transform data without switching tools. Mito includes a context-aware AI assistant that helps generate code, debug errors, and guide workflows directly inside Jupyter. Its spreadsheet layer supports familiar functions such as filters, pivot tables, and formulas, while automatically converting every action into production-ready Python code. This removes the need to manually translate spreadsheet logic into scripts. ... -
23
Code2Prompt
Convert codebases into structured prompts optimized for LLM analysis
...It analyzes a project directory, gathers relevant source files, and formats them into a single prompt that includes the source tree and code content. This approach helps developers quickly provide full project context to AI models without manually copying files or assembling prompts. code2prompt is built in Rust and focuses on performance, enabling fast traversal of large repositories while maintaining low resource usage. It also respects common project conventions such as .gitignore, ensuring that unnecessary files are automatically excluded from the generated prompt. ... -
24
verl-agent
Designed for training LLM/VLM agents via RL
...The framework supports multi-turn interactions between agents and their environments, allowing the system to receive feedback after each step and adjust its strategy accordingly. This step-wise interaction model makes it possible to train agents to operate in long-horizon scenarios where decisions depend on cumulative context and previous outcomes. Developers can configure memory modules that determine how historical information is stored and incorporated into each step of the reasoning process. -
25
LLM-Aided OCR Project
Enhances Tesseract OCR output using LLMs (local or API)
...The project addresses common OCR challenges such as distorted text, unusual fonts, historical documents, and complex layouts that often produce inaccurate results with standard OCR pipelines. The system first extracts raw text using OCR engines and then applies language models to analyze and correct recognition errors based on context. This AI-assisted correction process helps reconstruct missing characters, fix formatting mistakes, and produce more coherent text outputs. The project is particularly useful for digitizing historical documents, research papers, and scanned materials where traditional OCR often struggles. It also includes tools for processing batches of images or documents, enabling automated document digitization workflows.
