Gemma 4

“it loved to just stick Chinese and Korean characters into the text, sometimes without seeming to realize it had mixed languages”

“There simply haven't been any models that are better than Qwen 27B or Gemma 31B, which run comfortably in 64GB with big context.”

“They fail at even the most basic questions I throw at them, like 12B just now failed at answering how XGrabKey from Xlib is used.”

“I have banged on about this model here enough but I really agree that Gemma 4 12B is a candidate for the most impressive LLM of the year.”

“Reminds me of Gemma4 and the official (or at least popularly used around launch) Jinja templates being wrong and broken for tool calling.”

“Qwen 3.8 27B is only the second local model after Gemma 4 that managed to correctly reason through one of my private benchmarks.”

“there are tasks where Gemma is great, but on my camera watching, Qwen 3.6's MoE far exceeds either the Gemma 4 Dense or MoE”

“understands a lot of my references and influences, could recommend me my (still semi-obscure) favourite historical photographer”

“Only once have I tried something decent like Gemma 4 31B and Qwen 3.6 27B did I realize how incredibly useful they are.”

“Gemma 4 12B QAT at 4-bits is surely the smartest model in its size, but it shines at vision tasks rather than agentic tasks”

“better than anything you can run locally in ~8GB”

“locally, quantized in a 24GB scenario, Gemma 4 31B is better in my tests which are mostly reasoning or C programming related.”

“That's a tiny model. No LLM is going to be a perfect and repeatable judge, but a tiny 4B model is like plugging an RNG into this system.”

“Gemma 4 31b is a good model for other tasks but at least my personal experience is that Qwen outperforms in coding.”

“The hardware for 50 tokens per second with a four bit quantisation of Gemma 4 26B or the sparse Qwen 3.6 is not really that expensive”

“Gemma 4 12B can consistently select where to click on browser images given a minimal prompt, and do so very quickly.”

“Gemma 4 26B outperforms most 100-200B models that I've tested for reasoning and structured output.”

“I'm using Gemma-4 31B (gemma-4-31B-it-assistant.Q4_K_M.gguf) with llama.cpp to attribute quotations throughout chapters of my sci-fi novel.”

“I’ve been very impressed by it, even having run just about every other open weight model that would fit on my machine”

“Gemma 4 and Qwen 3.5 are well known for having fantastic general-purpose models in the 4B-9B range, but at the lower end they actually suck again”

“even "old" models like llama 3.2, gemma 2, qwen 2.5, are good, fast and low-memory”

“use enormously less tokens per output than other models, while still staying within arms reach of top benchmark performance”

“there are complex tasks that gemma will complete much much faster than qwen because it makes the right assumptions the first time”

“E4B was a reasonable stop along the line just to make sure the LoRA combined with the base model to produce coherent output”

“running the much cheaper, Gemma 4 four things where I don't need Max intelligence and running sonnet or opus for things where I need the intelligence and I can't really make the trade-off is been generally good”

“I did get one weird session where the model got confused and couldn't decide which tools actually existed in its inventory”

“Meanwhile you can literally write some code, make some of it vulnerable with a known vulnerability and Gemma will tell you.”

“which you can run on an off-the-shelf laptop, is around the same or even higher intelligence level as the SOTA models from 2 years ago”

“the release of open-weight Gemma models that can run on reasonable local hardware, and are not drastically worse than Anthropic flagships, is quite a punch”

“It is very fast (~100 t/s) and feels instant and very capable, and I have used Claude Code less and less these days.”

“Tool use, iterative refinement, managing large contexts, and reasoning outside of coding brought the scores back down to reality.”

“Impressive speed and quality for the small model with thinking off (E4B). Very slow for the big model with thinking turned on.”

“Gemma will be better at answering simple questions and doing basic automation, and codegen won't be it's strong suit.”

“it produced mostly good code for simple problems, but it sometimes hallucinated words in non-English scripts inside the code”

“First time in a long time that I have been impressed by a local model. Both speed (~38t/s) and quality are very nice.”

“They don't really have the structure of a short story, though the 20 GB model is more interesting and has two characters rather than just one character.”

“Gemma 4 is performing really well for me, and seems to hallucinate much less than other models I tried in this size range.”

“Google and Nvidia are significantly behind, not just on benchmarks but real-world performance like tool calling accuracy.”

“The model does call tools successfully giving sensible parameters but it seems to not picking the right ones in the right order.”

“I setup a pipeline for inference with OCR, full text search, embedding and summarization of land records dating back 1800s. All powered by the GGUF's you generate and llama.cpp.”