Running Qwen3-Coder-Next-NVFP4-GB10 with DFlash speculative decoding on SGLang. Tested on a Lenovo ThinkStation PGX (NVIDIA GB10 Grace Blackwell, 128 GB unified memory).
| Test | SGLang DFlash | vLLM DFlash+Marlin | Delta |
|---|---|---|---|
| Short code (307 tok) | 150 tok/s | 108 tok/s | +38% |
A pattern for building personal knowledge bases using LLMs.
This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.
Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.
We're thrilled to share a preview version of txtai agents. Inspired by the simplicity of frameworks like OpenAI Swarm, txtai agents are built on top of the Transformers Agent framework. This supports all LLMs txtai supports (Hugging Face, llama.cpp, OpenAI + Claude + AWS Bedrock via LiteLLM).
The following example shows how to create an agent with txtai. Agents will be available in the upcoming txtai 8.0 release (available now in the txtai GitHub repo now - follow #804 - feedback welcome).
