How did we all miss the release of AutoGen Studio 0.4.1.11? (incorporates new visual drag-and-drop interface for building agent workflows).

AutoGen Studio v0.4.1.7 released

Is there an “easy button” for running vLLM (Docker version) on a Windows 11 PC?

Finally figured it out - OpenWeb UI with your own, custom RAG back-end

Still haven't deployed an agent? This post will change that

What’s the Best PDF Extractor for RAG? I Tried LlamaParse, Unstructured and Vectorize

Anyone offering discount lawn service to their immediate neighbors since it’s just another zone add?

What’s recent open source LLMs have the largest context windows?

Does anyone know when the Yuka sweeper kit is going to be back in stock?

Am I the only person that didn't know this? Hashtag thing?

Open WebUI quietly releases 0.5.11, adding one of the best dev-focused features ever: Jupyter notebook support

A live look at the ReflectionR1 distillation process…

What version of DeepSeek can I run on a 4090?

I built and open-sourced a model-agnostic architecture that applies R1-inspired reasoning onto (in theory) any LLM. (More details in the comments.)

Twin stick games?

Talk me out of buying this 512GB/s Gen 5 NVMe RAID card + 4 drives to try to run 1.58bit DeepSeek-R1:671b on (in place of more RAM)

I really need to upgrade

Ollama 0.5.8 adds AVX-512 CPU acceleration and AVX2 for NVIDIA & AMD GPUs (pre release version available now).

All DeepSeek, all the time.

Open WebUI drops 3 new releases today. Code Interpreter, Native Tool Calling, Exa Search added

DeepSeek R1 ties o1 for first place on the Generalization Benchmark.

Open WebUI adds reasoning-focused features in two new releases OUT TODAY!!! 0.5.5 adds "Thinking" tag support to streamline reasoning model chats (works with R1) . 0.5.6 brings new "reasoning_effort" parameter to control cognitive effort.

“Boujee” is just a state of mind. Do the best with what you got.

Mistral Small 3 24b is the first model under 70b I’ve seen pass the “apple” test (even using Q4).

My PC 10 seconds after I typed “ollama run deepseek-r1:671b”: