#llm
Llm: 26 AI articles covering llm news, analysis, and research
Articles
How to Utilize OKF Efficiently to Enable Knowledge Exchange Among LLMs⭐10
Learn how OKF enables LLM knowledge exchange, cutting TTFT by up to 37.8% with shared token pointers across models.
Before Full Agentic RAG: Know How You Decide, and the Parsing⭐9
Before full agentic RAG, control parsing choices with an explicit dispatcher: plan, execute, and log each method for transparent, validated enterprise retrieval...
Liquid AI Unveils LFM2.5-2.6B: On-Device Agentic Model with 128K⭐7
Liquid AI's LFM2.5-2.6B brings on-device agentic AI with 128K context, tool calling, and open weights for edge deployment.
Unhealthy LLM Use Is More Common Than You Think⭐9
A recent study finds nearly 40% of regular users exhibit unhealthy LLM habits, from social replacement to anxiety, revealing problematic AI use is more common t...
Prompt, Context, Loop: The Three Engineering Layers Every RAG⭐10
Explore the three engineering layers of RAG systems: prompt, context, and loop. Learn how each layer works and why their evolution isn't a simple sequence.
Building CLI Agents with Python and Ollama⭐10
Build a Python CLI agent with Ollama to automate terminal tasks using local LLMs, tool calling, and code execution workflows.
AMD Releases Instella-MoE-16B-A3B: A Fully Open⭐7
AMD unveils Instella-MoE-16B-A3B, a fully open MoE LLM with 2.8B active parameters, trained on Instinct GPUs, offering complete transparency for research.
Coding Agents Don’t Need Bigger Context Windows — They Need a⭐10
Coding agents don't need bigger context windows—they need a compiler approach. This three-pass pipeline cuts prompt sizes by 69-74% in under 75ms.
Meet Token Saver: An Open-Source MCP Extension Using Local⭐8
Token Saver: an open-source MCP extension using local hybrid RAG to cut Claude PDF token costs by 90-99%.
As AI Content Floods the Internet, Pangram Raises $9M to Detect It⭐9
Pangram raises $9M to detect AI-generated text and images, launching Pangram 4 with over 99% accuracy in identifying machine-written content.