Awesome List Updates on Jun 21, 2024
3 awesome lists updated today.
🏠 Home · 🔍 Search · 🔥 Feed · 📮 Subscribe · ❤️ Sponsor
1. Awesome Azure Openai Llm
Microsoft Azure OpenAI relevant LLM Framework / Lucene based search engine with OpenAI Embedding
- LLMLingua (⭐4.3k): Compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss. [Jul 2023] LLMLingua-2 [Mar 2024]
- Autogen (⭐30k): Customizable and conversable agents framework ref [Mar 2023] Autogen Studio [June 2024 ]
Azure Reference Architectures / Lucene based search engine with OpenAI Embedding
- Guideline
- Grounding LLMs: Retrieval-Augmented Generation (RAG) [09 Jun 2023]
- Revolutionize your Enterprise Data with ChatGPT [09 Mar 2023]
- Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedback [07 Mar 2023]
- Azure OpenAI Design Patterns (⭐577): A set of design patterns using the Azure OpenAI service [May 2023]
- Azure AI Services Landing Zone (⭐84) / ref [24 Jul 2023]
- Security Best Practices for GenAI Applications (OpenAI) in Azure [16 Jan 2024]
- Authentication and Authorization in Generative AI applications with Entra ID and Azure AI Search [09 Jan 2024]
- Integrate private access to your Azure Open AI Chatbot [30 Nov 2023]
- Smart load balancing for OpenAI endpoints git (⭐75) [Jan 2024]
- An Introduction to LLMOps: Operationalizing and Managing Large Language Models using Azure ML [27 Aug 2023]
- Optimize Azure OpenAI Applications with Semantic Caching [09 Apr 2024]
- Azure OpenAI and Call Center Modernization [11 Apr2024]
- Azure OpenAI Best Practices Insights from Customer Journeys: LLMLingua, Skeleton Of Thought [12 Jun 2024]
Prompt Engineering / Prompt Template Language
- Skeleton Of Thought: Skeleton-of-Thought (SoT) reduces generation latency by first creating an answer's skeleton, then filling each skeleton point in parallel via API calls or batched decoding. [28 Jul 2023]
- NLEP (Natural Language Embedded Programs) for Hybrid Language Symbolic Reasoning: Use code as a scaffold for reasoning. NLEP achieves over 90% accuracy when prompting GPT-4. [19 Sep 2023]
Pruning and Sparsification / Llama Finetuning
- Pruning: The process of removing some of the neurons or layers from a neural network. This can be done by identifying and eliminating neurons or layers that have little or no impact on the network's output.
- Sparsification: A technique used to reduce the size of large language models by removing redundant parameters.
Other techniques and LLM patterns / Llama Finetuning
- Lamini Memory Tuning (⭐248): Mixture of Millions of Memory Experts (MoME). 95% LLM Accuracy, 10x Fewer Hallucinations. ref [Jun 2024]
Numbers LLM / GPT series release date
- Tokencost (⭐1.4k): Token price estimates for 400+ LLMs [Dec 2023]
Open-Source Large Language Models / GPT series release date
- Falcon LLM Apache 2.0 license [Mar 2023]
- StableVicuna First Open Source RLHF LLM Chatbot [Apr 2032]
- Alpaca: Fine-tuned from the LLaMA 7B model [Mar 2023]
- vicuna: 90% ChatGPT Quality [Mar 2023]
- Koala: Focus on dialogue data gathered from the web. [Apr 2023]
- dolly: Databricks [Mar 2023]
- Cerebras-GPT: 7 GPT models ranging from 111m to 13b parameters. [Mar 2023]
Agents: AutoGPT and Communicative Agents / Agent Design Patterns
- Agentic Design Patterns ref [Mar 2024]
- Reflection: LLM self-evaluates to improve.
- Self-Refine [30 Mar 2023]
- Reflexion [20 Mar 2023 ]
- CRITIC [19 May 2023]
- Tool use: LLM uses tools for information gathering, action, or data processing.
- Gorilla [24 May 2023]
- MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action [20 Mar 2023]
- Efficient Tool Use with Chain-of-Abstraction Reasoning [30 Jan 2024]
- Planning: LLM devises and executes multistep plans to reach goals.
- Multi-agent collaboration: Multiple AI agents collaborate for better solutions.
- Communicative Agents for Software Development [16 Jul 2023]
- AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation [16 Aug 2023]
- MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework [1 Aug 2023]
- Framework: Autogen (⭐30k) / LangGraph (⭐5.2k) / crewAI (⭐18k)
- Reflection: LLM self-evaluates to improve.
- Generate the code ref [Jun 2024]
2. Awesome Godot
Demos / Godot 4.x
- Kaleidoscope (⭐7) - A demo for interactive shaders in Godot.
3. Awesome Angular
Podcasts
Integrations / Google Developer Experts
- ngx-facebook-messenger (⭐4) - Ngx Facebook Messenger offers a lightweight alternative that optimizes user experiences by significantly reducing initial load size, ensuring smoother page performance, and enhancing overall website efficiency.
Free / Google Developer Experts
- carbon-angular-starter (⭐76) - A simple starter app for bootstrapping applications with Carbon.
Updating / Google Developer Experts
- ng16-dep-audit (⭐40) - Quickly check which dependencies stop you from upgrading!
- ngx-maintenance (⭐20) - Can't upgrade to Angular 16+ due to old View Engine dependencies? This automates the hassle so you can easily upgrade!
Animations / Google Developer Experts
- ngx-typed-writer (⭐6) - A Native Angular 2+ Typing Animation Library (Angular SSR and Angular Universal Friendly).
Charts / Google Developer Experts
- ngu-flow (⭐38) - Angular flow diagram library.
File Upload / Google Developer Experts
- ngx-filepond (⭐199) - A handy FilePond (⭐15k) adapter component for Angular.
Notifications / Google Developer Experts
- ngx-notifier (⭐4) - A Simple Notification Service for Angular applications.
Unspecified / Google Developer Experts
- carbon-components-angular (⭐525) - An Angular implementation of the Carbon Design System for IBM.
Misc Components / Google Developer Experts
- angular-paginator (⭐6) - Pagination Component for Angular applications.
- ngx-flamegraph (⭐57) - Flame graph for stack trace visualization written in Angular.
- angular-compiler-output (⭐15) - See the JS output of the Angular compiler for a given angular template.
RxJS / Google Developer Experts
- audit-debounce (⭐0) - RxJS operator to debounce and audit simultaenously.
- Prev: Jun 22, 2024
- Next: Jun 20, 2024