otAI/RAGatouille|RAGatouille]] : bridging the gap between state-of-the-art research and alchemical RAG... t and accurate retrieval model, enabling scalable BERT-based search over large text collections in ten... at leaf level, ascends tree only if confidence is below threshold
<code>
query
→ multi-query expans... icles
* [[https://aimultiple.com/rag-frameworks|Benchmark 5 RAG frameworks: LangChain, LangGraph, Ll
====== Model bench ======
Avec OpenCode et le prompt "hello".
gemma-4-26B-A4B-it-Q4_K_M
''~/Code/br...
0.35.423.655 I srv load: - looking for better prompt, base f_keep = -1.000, sim = 0.000
0.3...
0.39.629.094 I srv load: - looking for better prompt, base f_keep = -1.000, sim = 0.000
0.3...
0.53.859.929 I srv load: - looking for better prompt, base f_keep = -1.000, sim = 0.000
0.5
e availability of each administrative team member between 9h and 18h.)
* Identify common available t... ility (Request each member’s available time slots between 9h and 18h.)
* Determine common slot (Anal... ers (Send a reminder to all participants 24 hours before the meeting.)
* Prepare agenda (Draft and d... dministrative staff asking for their availability between 9h and 18h today. Include options for each h
is perso]]
* Un peu de [[informatique:ai_lm:gpu_bench|GPU bench]]
* [[/informatique:ai_lm:ai_image|AI Image]] pour la génération d'image avec un LM
... //github.com/MariaDB/mcp
* [[https://github.com/benborla/mcp-server-mysql|benborla/mcp-server-mysql]] (Node.js), 22 contributors, 1.1k stars, 281 forks,
Client Protocol (ACP) standardizes communication between code editors/IDEs and coding agents and is s... rch.aimultiple.com/rag-frameworks/|RAG frameworks benchmark results: LangChain vs LangGraph vs LlamaIn... ns'' + un gros ''System Prompt'' il ne reste plus beaucoup de ''Context Window'' pour le reste. Aussi
