Florian Brand, Prime Intellect research engineer, adopts Gemma 4 E4B 6-bit quantized as his primary local Mac LLM

💎 @googlegemma

Gemma 4 E4B 6bit is now the local model of my choice and loaded 24/7 on my Mac (using @lmstudio), replacing Qwen3, 3.5 4B after ~9 months of usage

What an insane model, congrats @GoogleDeepMind 🤠

@xeophon @yacineMTB @lmstudio @GoogleDeepMind Wouldn’t qwen 9b be nicer?

@xeophon @lmstudio @GoogleDeepMind Try the uncensored version, it's so much better imo

@xeophon @lmstudio @GoogleDeepMind what are ur usecases? "rewrite", "summarize", "translate," or something bigger in scope and harder by nature?

@xeophon @lmstudio @GoogleDeepMind Wouldn't the 4Bit QAT be better than a 6Bit PTQ

@RaghavKoch19380 @lmstudio @GoogleDeepMind The QAT are GGUF only afaik

@ignis_code @lmstudio @GoogleDeepMind M4 Max + 64 GB, model uses 7 GB

@xeophon @lmstudio @GoogleDeepMind There are compressed tensor versions or something available for vLLM etc i think. check their huggingface QAT folder.

@xeophon @lmstudio @GoogleDeepMind what are you using it for?

@0xgeorge @yacineMTB @lmstudio @GoogleDeepMind License

@xeophon @lmstudio @GoogleDeepMind Is this also over Gemma 4 12B? https://blog.google/innovation-and-ai/technology/developers-tools/introducing-gemma-4-12B/

@xeophon @wambosec @lmstudio @GoogleDeepMind 64 gb and you use Gemma4 e4b ??? Bro at least use gemma4 12b

@xeophon @lmstudio @GoogleDeepMind 어느정도의 VRAM을 사용하시나요?

@xeophon @lmstudio @GoogleDeepMind yes, even i have one model always loaded on my system for assistance while building stuff or solving any problems.

i think people who can use small 4-9B models to build stuff can actually be called coders.

@xeophon @yacineMTB @lmstudio @GoogleDeepMind Why not LFM 2.5 at 8bit for just an extra gb?

@xeophon @lmstudio @GoogleDeepMind mac specs?

@xeophon @lmstudio @GoogleDeepMind Are you using it for the privacy considerations, Xeo?

@xeophon @lmstudio @GoogleDeepMind What context window are you using?

@xeophon @lmstudio @GoogleDeepMind Why?