Local models need focus and polish.
Inkling: Our Open-Weights Model
Our first open-weights model: multimodal, Mixture-of-Experts, with controllable reasoning effort. Available to fine-tune on Tinker.
Better Models: Worse Tools
About an aggravating tool-calling regression in newer Claude models.
Why and How to Run Local Models in Zed
From the Zed Blog: You can run local AI models in Zed to get better performance and control over your data. Here's how.
Open Responses with local models via LM Studio
Update to LM Studio 0.3.39 for Open Responses support
Introducing Model Council
Today we are launching Model Council, a multi-model research feature that brings several models together for one answer.
Open models are decelerationist - Erlend’s notes
people and planet need open models to win
Running local models is good now
Local agentic coding has gotten great over the past few months
Recursive Language Models
We study allowing large language models (LLMs) to process arbitrarily long prompts through the lens of inference-time scaling. We propose Recursive Language Models (RLMs), a general inference paradigm that treats long prompts as part of an external environment and allows the LLM to programmatically examine, decompose, and recursively call itself over snippets of the prompt. We find that RLMs can successfully process inputs up to two orders of magnitude beyond model context windows and, even for shorter prompts, dramatically outperform the quality of vanilla frontier LLMs and common long-context and coding scaffolds (e.g., on GPT-5 by a median across the evaluated benchmarks of $26\%$ against compaction, $130\%$ against CodeAct with sub-calls, and $13\%$ against Claude Code) across four diverse long-context tasks while having comparable cost. At a small scale, we post-train the first model around the RLM. Our model, RLM-Qwen3-8B, outperforms the underlying Qwen3-8B model by $28.3\%$ on average and even approaches the quality of vanilla GPT-5 on three long-context tasks. Code is available at https://github.com/alexzhang13/rlm.
Muse Glimmer: Meta’s 30B Model Built for Efficient Inference
Inside Meta’s 30B local reasoning model and its tiny KV cache
Ollama's new engine for multimodal models· Ollama Blog
Ollama now supports new multimodal models with its new engine.
Own your AI with local models and open protocols | Tiles Blog
A Local-First Conf talk about local models, open protocols, and user-owned AI.
Project Cambria: Translate your data with lenses
Changing schemas in distributed software is hard. Could adopting bidirectional lenses help?
> Making the models smarter doesn't solve the problem. It makes the problem harder to see. So many relatable sentences here.
M Berk
I found this article so, so helpful at explaining why slogging through is the best way to learn (and so much more): ergosphere.blog/posts/the-machines-are-fine/
