Search Results for "expert"
Sort By:
Showing 323 open source projects for "expert"
View related business solutions-
Ship Agents FasterGemini Enterprise Agent Platform lets you rapidly build, scale, govern and optimize production-ready agents grounded in your organization's data. The platform enables developers to build custom or pre-built agents for virtually any use case. New customers get $300 in free credits.
-
$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
DeepEP
DeepEP: an efficient expert-parallel communication library
DeepEP is a communication library designed specifically to support Mixture-of-Experts (MoE) and expert parallelism (EP) deployments. Its core role is to implement high-throughput, low-latency all-to-all GPU communication kernels, which handle the dispatching of tokens to different experts (or shards) and then combining expert outputs back into the main data flow. Because MoE architectures require routing inputs to different experts, communication overhead can become a bottleneck — DeepEP addresses that by providing optimized GPU kernels and efficient dispatch/combining logic. ... -
2
EPLB
Expert Parallelism Load Balancer
...It uses policies like hierarchical load balancing (grouped experts placed at node and then GPU level) and global load balancing depending on configuration. The logic is implemented in eplb.py and supports predicting placements given estimated expert usage weights. EPLB aims to reduce hot-spotting and ensure more uniform usage of compute resources in large MoE deployments. -
3
Halo
Framework built by White Circle for training large language
...It supports pretraining, supervised fine-tuning, preference optimization, distillation, and fully asynchronous multi-turn reinforcement learning. The framework trains Hugging Face models directly, so checkpoints remain compatible with standard from_pretrained workflows. Distributed execution supports expert, context, tensor, and expert-tensor parallelism alongside FSDP2. 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. -
4
-
Fully Managed MySQL, PostgreSQL, and SQL ServerCloud SQL handles your database ops end to end, so you can focus on your app.
-
5
Damn Vulnerable GraphQL Application
Vulnerable implementation of Facebook's GraphQL technology
...DVGA has numerous flaws, such as Injections, Code Executions, Bypasses, Denial of Service, and more. See the full list under the Scenarios section. A public Postman collection is also available to replay solutions to the challenges. DVGA supports Beginner and Expert level game modes, which will change the exploitation difficulty. -
6
Council of High Intelligence
18 AI personas deliberate your hardest decisions
Council of High Intelligence is an AI deliberation framework that lets multiple expert-style personas analyze hard decisions together. Instead of relying on one model response, it creates structured disagreement between different intellectual roles such as Socrates, Feynman, Kahneman, Torvalds, Aristotle, and others. The project supports multi-round reasoning, cross-examination, dissent quotas, novelty checks, and model diversity across different providers. -
7
MagiCrew
The first open-source all-in-one AI productivity platform
MagiCrew is an open source enterprise AI agent platform that combines assistants, expert agents, workflow automation, messaging, and collaborative productivity tools. It centralizes organizational data and knowledge so AI capabilities can be shared across teams. Administrators can set budgets by department, user, or task to control model spending. Sandboxed execution and human approval gates help contain risky automated actions. -
8
Colibrì
Run frontier MoE models on hardware you already own
Colibrì is an open-source inference engine designed to run frontier-scale mixture-of-experts (MoE) models on consumer and heterogeneous hardware. It treats VRAM, system RAM, and NVMe storage as a unified memory hierarchy, dynamically placing and streaming model weights where they can be accessed most efficiently. This architecture enables users to run models ranging from 7 billion to 2.8 trillion parameters without requiring the entire model to fit in expensive GPU memory. Written primarily... -
9
Preguntas de entrevista para React
Questions about React for job interviews
Preguntas de entrevista para React is a comprehensive Spanish-language resource containing typical React interview questions organized from beginner to expert level. The repository is not just a list of questions: it includes detailed explanations in Spanish, making it useful as both an interview prep guide and a deeper learning reference.The content is structured in sections (beginner, intermediate, expert), covering everything from “What is React?” and JSX basics to advanced topics such as hydration, server components, useTransition, useSyncExternalStore, server actions, streaming rendering APIs, and new React 19 features. ... -
Build Securely on Azure with Proven FrameworksMoving to the cloud brings new challenges. How can you manage a larger attack surface while ensuring great network performance? Turn to Fortinet’s Tested Reference Architectures, blueprints for designing and securing cloud environments built by cybersecurity experts. Learn more and explore use cases in this white paper.
-
10
Ring
Ring is a reasoning MoE LLM provided and open-sourced by InclusionAI
Ring is a reasoning Mixture-of-Experts (MoE) large language model (LLM) developed by inclusionAI. It is built from or derived from Ling. Its design emphasizes reasoning, efficiency, and modular expert activation. In its “flash” variant (Ring-flash-2.0), it optimizes inference by activating only a subset of experts. It applies reinforcement learning/reasoning optimization techniques. Its architectures and training approaches are tuned to enable efficient and capable reasoning performance. Reasoning-optimized model with reinforcement learning enhancements. ... -
11
Tilt
Define dev environment as code for microservice apps on Kubernetes
...Our engine starts the whole app and runs automated rebuilds as you edit in your IDE. Get a continuous feedback loop with your logs, broken builds, and runtime errors. Work with Kubernetes without needing to be an expert. And if you are an expert, no more 20 questions with kubectl. Tilt responsively handles the tedious and repetitive parts of your workflow and gives you peripheral vision so you find errors faster. Recapture the magic of hacking with immediate feedback. -
12
i.am.ai
Roadmap to becoming an Artificial Intelligence Expert in 2022
i.am.ai is a structured educational guide that maps out the knowledge areas and technologies required to become an artificial intelligence or machine learning expert. The project presents visual charts that outline multiple career paths such as data scientist, machine learning engineer, and AI specialist, helping learners understand what to study and in what order. It was originally created to train internal employees but was released publicly to support the broader community. The roadmap emphasizes foundational skills like mathematics, programming, and data handling before progressing into deep learning and specialized domains. ... -
13
Tencent-Hunyuan-Large
Open-source large language model family from Tencent Hunyuan
Tencent-Hunyuan-Large is the flagship open-source large language model family from Tencent Hunyuan, offering both pre-trained and instruct (fine-tuned) variants. It is designed with long-context capabilities, quantization support, and high performance on benchmarks across general reasoning, mathematics, language understanding, and Chinese / multilingual tasks. It aims to provide competitive capability with efficient deployment and inference. FP8 quantization support to reduce memory usage... -
14
CoStrict
Strict AI coder for enterprises, quality first
...The system integrates repository-wide analysis using retrieval-augmented generation, allowing it to understand large codebases and provide context-aware suggestions, reviews, and modifications. It also incorporates multi-agent or multi-expert verification strategies, ensuring that generated code is validated from multiple perspectives before being accepted. -
15
Ling
Ling is a MoE LLM provided and open-sourced by InclusionAI
Ling is a Mixture-of-Experts (MoE) large language model (LLM) provided and open-sourced by inclusionAI. The project offers different sizes (Ling-lite, Ling-plus) and emphasizes flexibility and efficiency: being able to scale, adapt expert activation, and perform across a range of natural language/reasoning tasks. Example scripts, inference pipelines, and documentation. The codebase includes inference, examples, models, documentation, and model download infrastructure. As more developers and researchers engage with the platform, we can expect rapid advancements and improvements, leading to even more sophisticated applications. ... -
16
UCCL
UCCL is an efficient communication library for GPUs
...UCCL is designed to work with heterogeneous hardware environments, allowing GPUs from different vendors and network interfaces to communicate efficiently without vendor lock-in. The system also supports specialized workloads such as reinforcement learning weight transfers, key-value cache sharing, and expert parallelism for mixture-of-experts models. Its architecture emphasizes flexibility and extensibility so that developers can implement custom communication protocols tailored to specific machine learning workloads. -
17
ImageReward
[NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences
ImageReward is the first general-purpose human preference reward model (RM) designed for evaluating text-to-image generation, introduced alongside the NeurIPS 2023 paper ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation. Trained on 137k expert-annotated image pairs, ImageReward significantly outperforms existing scoring methods like CLIP, Aesthetic, and BLIP in capturing human visual preferences. It is provided as a Python package (image-reward) that enables quick scoring of generated images against textual prompts, with APIs for ranking, scoring, and filtering outputs. ... -
18
Orange Data Mining
Orange: Interactive data analysis
Open source machine learning and data visualization. Build data analysis workflows visually, with a large, diverse toolbox. Perform simple data analysis with clever data visualization. Explore statistical distributions, box plots and scatter plots, or dive deeper with decision trees, hierarchical clustering, heatmaps, MDS and linear projections. Even your multidimensional data can become sensible in 2D, especially with clever attribute ranking and selections. Interactive data exploration for... -
19
Koishi
Cross-platform chatbot framework made with love
...It lets users create chatbots for platforms such as QQ, Telegram, Discord, Feishu, and other messaging services. The project includes a web console that makes setup, monitoring, plugin installation, and behavior control easier for non-expert users. Its ecosystem includes thousands of official and community plugins for adapters, databases, storage, web panels, state handling, and business features. Developers get a TypeScript-based framework with strong typing, unit-tested core behavior, and hot reloading for plugin development. Koishi works well for both lightweight helper bots and larger interactive applications that need a stable plugin architecture. -
20
Swift Concurrency Agent Skill
Add expert Swift Concurrency guidance to your AI coding tool
Swift Concurrency Agent Skill is an open-source “agent skill” designed to give AI coding assistants deep expertise in Apple’s Swift Concurrency model, including async/await, structured concurrency, task groups, actors, and thread safety. It is formatted according to the Agent Skills specification so that tools like Claude Code, Cursor, Copilot, and other LLM-powered systems can load it and apply guidance when relevant. The skill codifies practical best practices for writing efficient, safe,... -
21
bolt.diy
Prompt, run, edit, & deploy full-stack web applications using any LLM
bolt.diy is an open-source platform that allows you to easily create, run, edit, and deploy full-stack web applications using a variety of large language models (LLMs). It supports popular models like OpenAI, Anthropic, Ollama, OpenRouter, Gemini, LMStudio, Mistral, xAI, HuggingFace, DeepSeek, and Groq, and provides the flexibility to integrate additional models through the Vercel AI SDK. Whether you’re experimenting with pre-built models or developing custom AI-driven applications, bolt.diy... -
22
Wan2.2
Wan2.2: Open and Advanced Large-Scale Video Generative Model
Wan2.2 is a major upgrade to the Wan series of open and advanced large-scale video generative models, incorporating cutting-edge innovations to boost video generation quality and efficiency. It introduces a Mixture-of-Experts (MoE) architecture that splits the denoising process across specialized expert models, increasing total model capacity without raising computational costs. Wan2.2 integrates meticulously curated cinematic aesthetic data, enabling precise control over lighting, composition, color tone, and more, for high-quality, customizable video styles. The model is trained on significantly larger datasets than its predecessor, greatly enhancing motion complexity, semantic understanding, and aesthetic diversity. ... -
23
WASTE
Run the full 2.78-trillion-parameter Kimi K3 model
WASTE is an embeddable C inference engine for running extremely large mixture-of-experts models when the weights exceed available RAM. It keeps the shared model trunk in memory and streams only the experts selected for each token from fast NVMe storage. A bounded cache reuses recently needed experts, while lookahead routing begins disk reads before the next layer requires them. Its main target is the full 2.78-trillion-parameter Kimi K3 model, including multimodal image input. The engine has... -
24
GitHub Actions for DigitalOcean
GitHub Actions for DigitalOcean - doctl
...Powerful and production-ready, our cloud platform has the solutions that devs like you need to succeed, whether you're building world-changing AI apps, running a side project, or building a business. GPU solutions for everyone—novice to expert. Run training and inference, process large data sets and complex neural networks, and deploy high-performance computing clusters. -
25
Mobile Verification Toolkit
Helps with conducting forensics of mobile devices
...Using it requires understanding the basics of forensic analysis and using command-line tools. This is not intended for end-user self-assessment. If you are concerned with the security of your device please seek expert assistance. Compare extracted records to a provided list of malicious indicators in STIX2 format. Generate JSON logs of extracted records, and separate JSON logs of all detected malicious traces.
