AgentCPM-Explore🔥 on device agent foundation model released by OpenBMB openbmb/AgentCPM-Explore ✨ 4B - Apache2.0 ✨ Supports 100+ multi-turn environment interactions with search + verification ✨ Full training/inference stack is openly shared as well
The models come in Thinking and Instruct versions and utilize a new architecture, allowing it to have ~10x faster inference than Qwen32B. 💜 Step-by-step Guide: https://docs.unsloth.ai/models/qwen3-next
I run 20 AI coding agents locally on my desktop workstation at 400+ tokens/sec with MiniMax-M2. It’s a Sonnet drop-in replacement in my Cursor, Claude Code, Droid, Kilo and Cline peak at 11k tok/sec input and 433 tok/s output, can generate 1B+ tok/m.All with 196k context window. I'm running it for 6 days now with this config.
Today max performance was stable at 490.2 tokens/sec across 48 concurrent clients and MiniMax M2.
Running large language models efficiently is more than just raw GPU power. The latest guide breaks down the essential math to determine if your LLM workload is compute-bound or memory-bound.
We apply these principles to a real-world example: Qwen's 32B parameter model on the new NVIDIA RTX PRO 6000 Blackwell Edition.
In this guide, you will learn how to:
Calculate your GPU's operational intensity (Ops:Byte Ratio) Determine your model's arithmetic intensity Identify whether your workload is memory-bound or compute-bound
Join us on site October 22-23 to see how Arm empowers developers to build and deploy AI applications with ease using PyTorch and ExecuTorch. Learn about the latest AI technologies from Arm and our ecosystem while expanding your professional network alongside like-minded AI engineers.
a senior engineer at google just dropped a 400-page free book on docs for review: agentic design patterns.
the table of contents looks like everything you need to know about agents + code: > advanced prompt techniques > multi-agent patterns > tool use and MCP > you name it
This is what efficient AI looks like: Gemma 3n just dropped - a natively multimodal model that runs entirely on your device. No cloud. No API calls.
🧠 Text, image, audio, and video - handled locally. ⚡️Only needs 2B in GPU memory to run 🤯 First sub-10B model to hit 1300+ Elo ✅ Plug-and-play with Hugging Face, MLX, llama.cpp, and more.
Plus: Multilingual out of the box (140+ languages), fine-tune in a free Colab notebook.
🧠 We just implemented Andrej Karpathy's "third paradigm" for LLM learning!
System Prompt Learning (SPL) enables LLMs to automatically learn problem-solving strategies from experience, rather than relying on static prompts.
🚀 How it works: Your LLM builds a database of effective strategies, selects the best ones for each problem, and refines them over time based on success rates.
The best part? All strategies are human-readable and the system gets progressively better at problem types you use frequently.
✨ Key benefits: 🔄 Cumulative learning over time 📖 Transparent, inspectable strategies 🔌 Works with any OpenAI-compatible API ⚡ Simple integration: just add "spl-" prefix to your model
Built as an open-source plugin in optillm. After 500 queries, our system developed 129 strategies and refined 97 of them!
This feels like a genuine step toward AI that learns from experience while staying completely interpretable.
✨Skywork OR1-Math-7B > Optimized for math reasoning ✨Skywork-OR1-7B-preview > Excels in math & coding ✨Skywork-OR1-32B-preview > Matches Deepseek-R1 on math (AIME24/25) and coding (LiveCodeBench)
Released under the Apache 2.0 license 🥳 Final version coming in 2 weeks!
As we always use Transformers, it's helpful to understand RoPE—Rotary Position Embedding. Since token order matters, RoPE encodes it by rotating token embeddings based on their position, so the model knows how to interpret which token comes first, second, and so on.
Here are 8 types of RoPE that can be implemented in different cases:
4. Multimodal RoPE (MRoPE) -> Qwen2.5-VL Technical Report (2502.13923) Decomposes positional embedding into 3 components: temporal, height and width, so that positional features are aligned across modalities: text, images and videos.
8. XPos (Extrapolatable Position Embedding) -> https://huggingface.co/papers/2212.10 Introduces an exponential decay factor into the rotation matrix, improving stability on long sequences.
