Repo Prompt and Ollama
For a question about using Ollama in the same manner as o1 works (i.e., returning the Whole or Diff format ready for merging), the repository owner provided the following response: Ollama models…
Repo Prompt
Quite nice application for working with files and LLMs: Repo Prompt. I have done two tests so far, and I’m sold. The tests, of course, were super simple apps, but I found…
BlenderGPT, MoneyPrinterV2
This is the next area where AI aids designers – 3D modeling. This page offers free access to create meshes from text or images. Additionally, there are models available on Hugging Face…
Llama 3.3, Ollama structured output
Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model. This model also requires less VRAM, e.g. works well on M4 64GB with a speed of 10 tokens/s. https://ollama.com/library/llama3.3 Right…
Chat from LangChain
This is a significantly different RAG we used to think of: https://chat.langchain.com/. Take a look at the sample questions and observe how agents from LangChain and LangGraph utilize them to construct responses….
Genie 2, DeepThought-8B, snowflake-arctic-embed2
Google has recently released Genie 2, their most advanced large-scale foundation model capable of generating consistent and playable worlds for a minute. This advancement could be particularly beneficial for game developers, but…
QwQ-32B-Preview
QwQ-32B-Preview: An Experimental AI Model The Qwen Team’s experimental research model, QwQ-32B-Preview, showcases promising analytical abilities but with several limitations: Language Mixing and Code-Switching: May switch between languages unexpectedly. Recursive Reasoning Loops:…
OpenWebUI + Ollama
Models Models keep information about existing models, allow us to modify them by adding functions, tools, system prompts, etc., or even create custom models on existing LLMs like Llama 3.x. It’s worth…
DeepSeek-R1-Lite-Preview Challenges US AI Models
DeepSeek-R1-Lite-Preview is Now Live DeepSeek-R1-Lite-Preview is a new model from China, which is a direct competitor to the best US models like OpenAI’s o1 and Claude 3.5 Sonnet, beating them in some…