🤖 Artificial Intelligence Jun 8, 2026 · Chew Loong Nian - AI ENGINEER

Google Shrank Gemma 4 by 72% and Unsloth Fixed the 4-Bit Bug Nobody Else Caught on One 4090, and 4-Bit Shouldn’t Be This Good

Towards AI
View Channel →
Google Shrank Gemma 4 by 72% and Unsloth Fixed the 4-Bit Bug Nobody Else Caught on One 4090, and 4-Bit Shouldn’t Be This Good
Source ↗ 👁 3 💬 0
Last Updated on June 8, 2026 by Editorial Team Author(s): Chew Loong Nian – AI ENGINEER Originally published on Towards AI. Google Shrank Gemma 4 by 72% and Unsloth Fixed the 4-Bit Bug Nobody Else Caught on One 4090, and 4-Bit Shouldn't Be This Good A 26-billion-parameter model has no business fitting in 15GB of memory and spitting out 193 tokens a second on a single consumer GPU. That is laptop-and-gaming-rig territory, not a datacenter. Yet that is exactly what Google’s new Gemma 4 QAT checkpo

Comments (0)

Sign in to join the discussion

More Like This

The Verified Identity Agent Bridge
Towards AI · Jun 18, 2026
You Can’t Prompt Your Away Your LLM Problems
Towards AI · Jun 18, 2026
The Free Agent Trap
Towards AI · Jun 18, 2026
📰
Your Agentic Loop Will Drift. Here Is the KL Divergence Equation That Measures How Far It Has Wandered From Its Original Instruction.
Towards AI · Jun 18, 2026
Beyond Chat: Processing Images, PDFs, and Documents with the OpenAI Adapter in Oracle Integration Cloud
Towards AI · Jun 18, 2026
Building AI Agents in Rust — part 3
Towards AI · Jun 18, 2026