Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-05-11 · source-backed
This model combines CogViT visual encoder with GLM-0.5B in under a billion parameters and beats everything on document OCR. Deploys on vLLM, SGLang, and Ollama. If you're running a document processing pipeline, this should be your first evaluation target.
Each link below shares sources, entities, or timing with this story.
For about a year, "run your agent locally" meant accepting a model that couldn't reliably call a tool twice in a row. That excuse is gone. Meta Superintelligence Labs published Muse Glimmer today: a 29.6B dense causal transformer, 52 layers, 6,656 hidden dim, with a ~1.8B ViT-...
Google released open-source Multi-Token Prediction (MTP) drafters for the Gemma 4 model family. The concept: pair a heavy target model (Gemma 4 31B) with a lightweight drafter that predicts several future tokens in parallel. The target model verifies the predictions in a singl...
Claims up to 2x faster generation, and adds fine-tuning of both MoE models on text or image datasets on Apple Silicon via MLX (release). Follow-up turns in long Qwen chats on Mac are reported up to 30x faster, MLX models now use full context size, and GLM-5.3 MLX fine-tunes ex...
The July 6 release delivers nearly 90% faster Gemma 4 token generation through multi-token prediction with automatic draft-length tuning, on by default, output-preserving, no config (Ollama). It also adds MLX-engine support for more model families and flash attention for older...
I've spent the last year assuming that if I wanted real agentic coding quality, I paid for a closed model. That assumption took a hit on June 1. MiniMax shipped M3 with a new sparse-attention architecture (they call it MSA) that handles up to 1M tokens at roughly 9x prefill an...
Google DeepMind released Gemma 4 on April 2 with four model sizes (E2B, E4B, 26B MoE, 31B Dense) under Apache 2.0. Multimodal (text, vision, audio). 256K context. Native thinking and tool-calling optimized for agentic workflows. Day-zero ecosystem support across vLLM, llama.cp...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.