Fetching from the wire…
01
02
03
04
05
06
07
08
Models2026-06-08 · source-backed
Google released Gemma 4 12B June 3 under clean Apache 2.0, native multimodality, up to 256K context on larger variants, with the 31B reportedly at 85.2% MMLU Pro. Two days later came QAT versions optimized for mobile and laptop hardware. The 12B size targets the single-GPU sweet spot, and QAT preserves accuracy at lower bit-widths, which makes local agentic workloads actually practical instead of aspirational.
Each link below shares sources, entities, or timing with this story.
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...
The models are good. The license is the real story. Google released Gemma 4 on April 2 with four variants: E2B, E4B, 26B MoE, and 31B Dense. All built on the Gemini 3 architecture. The 31B Dense variant claimed #3 on Arena AI's text leaderboard, beating models 20x its size. Th...
Google dropped Gemma 4 and it's not incremental. The 31B dense model ranks #3 on Arena AI with an ELO of 1,452, scores 85.2% on MMLU Pro, 89.2% on AIME 2026, and 80.0% on LiveCodeBench v6. It outperforms models 20x its size. Under Apache 2.0. At $0.20 per run. Only Opus 4.6 an...
Amid a week of pricing and commerce stories, here's hard tech you can actually download. Google released DiffusionGemma on June 10, a 26B-parameter Mixture-of-Experts model (3.8B active) that generates text by diffusion instead of left-to-right decoding. The architecture is th...
Google released Gemma 4 on April 2 with four model variants: E2B, E4B, 26B MoE, and 31B Dense. The license change is the first thing worth noting. Every previous Gemma had restrictions that made lawyers nervous. Gemma 4 is Apache 2.0. Full stop. Use it in any product, any way...
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...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.