Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Gemma 4 12B: A unified, encoder-free multimodal model

blog.google

Post reply on HN
Print view View original

371–380 of 421 posts

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#371

Post by dofm »

hypfer wrote: post #338
emmelaich wrote: post #311

Earlier quoted context omitted.

You may like https://www.llmfit.org/ (not recommendation, I've not used it .. yet)

Just tried it and honestly it's a terrible experience lacking any sort of intent or reason. Which is unsurprising in the AI space. You get a wall of text showing you various random fine-tuned models by random people, and that is basically it. Actual sane default requirements like "just give me the normal AI labs", "please filter for dense only" and "I want this exact context size at this quant" are not part of the to…

I have found these things to be fully exasperating, to be honest, even though I am seeking information about a pretty "known" machine — a 64GB M1 Max MBP.

(Honestly I think Apple's "AI push" could do worse than just focus on a curated model library, a couple of Apple-standard Gemini distillations, an OS-level model manager and some sort of tweak of their containers system to do what Docker's sbx does. They could demystify a lot of this shit.)

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#372

Post by borissk »

senko wrote: post #120

I ran the Q4 quant (used with llama.cpp) though my "minesweeper" vibe-coding benchmark: https://senko.net/vibecode-bench/2026/minesweeper-gamma-4-12... The result is decent, but it had a few bizzare/trivial syntax errors I had to fix manually: it would do an extra closing bracket or paren a few times, and wanted to separate function definitions with comma. Not sure what that was about, but otherwise the output run ju…

We are really getting close to singularity - the pace of LLM improvement is constantly accelerating.

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#373

Post by robgough »

fittingopposite wrote: post #328
robgough wrote: post #81

Earlier quoted context omitted.

I've got a home-built dictation app that uses a local model to clear up the text and fix grammar. It was super easy to build. I’m extending it to capture meeting notes and summarise too. All on-device. I saw a little app the other day, I think someone posted on here, that looks at your screenshot and renames the file based off the contents of the file. There's tons of little examples like that. For a lot of use cases…

That's a great user case. Am sorry using parakeet but sometimes it garbles up things. Can you open source it?

Mac only I'm afraid, but I already did [1]. Packaged it too as I figured it might be useful to others [2] (and I'd want to install it on machine's that I might not have Xcode on)

[1]: https://github.com/robgough/dictator [2]: https://dictator.robgough.net

I spent the best part of a couple weeks making improvements and tidying up the UI, but to actually get something working was essentially only a couple of prompts.

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#375

Post by UncleOxidant »

senko wrote: post #120

I ran the Q4 quant (used with llama.cpp) though my "minesweeper" vibe-coding benchmark: https://senko.net/vibecode-bench/2026/minesweeper-gamma-4-12... The result is decent, but it had a few bizzare/trivial syntax errors I had to fix manually: it would do an extra closing bracket or paren a few times, and wanted to separate function definitions with comma. Not sure what that was about, but otherwise the output run ju…

I've heard the assertion that the Gemma 4 models don't do well with lower quantization. I wonder if the "bizzare/trivial" syntax errors would go away at Q8?

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#376

Post by staticman2 »

0xbadcafebee wrote: post #288
asim wrote: post #137

We are now entering the closed loop game. Google doesn't need anyone else to accelerate their models. This is their bread and butter. I'm both shocked but also not surprised that they continue to develop such efficiencies. Honestly it's like silicon and CPU architecture advancement. We kept shrinking it and shrinking it and it kept getting more and more powerful and here we are with AI and it's only going to be 100x…

1996 didn't look that different than today, in the US anyway. Biggest difference, besides the electric cars, is everybody has a phone but nobody uses it to talk to people.

I agree the last 30 years in the U.S. hasn't changed all that much due to tech.

It's probably true that phones and social networks have altered the way people think, but not necessarily in a way that's qualitatively different from cable TV changing the way people in the 90s thought compared to people in the 60s...

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#377

Post by staticman2 »

__mharrison__ wrote: post #252
mchusma wrote: post #139

Earlier quoted context omitted.

I think its even more puzzling because you can't even run Gemma 31b on google cloud, they only let you test it with a rate limit. No way (I can find) to actually pay them to use it. We saw great results in our usecase using google direct. Moved to Openrouter because google wouldn't let us use it beyond a test. Then Openrouters performance looked worse, not sure if there was a quantized version or something. So we ins…

I'm using it for one of my use cases (ocr) on openrouter right now.

I tested Gemma 4 31b for OCR and it's very good at it. This makes sense because I also get the best OCR results from Gemini compared to Claude or ChatGPT in my use case.

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#378

Post by staticman2 »

petercooper wrote: post #132

Its image processing is terrible. I ran several tests against it against Qwen 3.5 0.8b (yes, 7% the size) and Qwen beat it every time with Gemma often getting things entirely wrong. I even gave it a plain image saying "This is a test" and it thought for 6 minutes trying to analyze it and failed. Qwen 3.5 0.8b confidently got it in under a second . It may be that the Q6 quant I got is borked (or my LM Studio is), but…

Test it on a professional inference provider to rule out trouble on your end.

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#379

Post by selicos »

dirkg wrote: post #326
0xbadcafebee wrote: post #206

Earlier quoted context omitted.

It was almost certainly not trained for coding, as it's got both audio and vision input, is only 12B, and nowhere in the announcement is coding mentioned. It will likely not have good performance on coding in general, compared to other small models like Qwen 3.6 35B A3B, Gemma 4 26B A4B, Nvidia Nemotron 3 Nano 30B-A3B, gpt-oss-20b. For 16GB laptops, Qwen 3.5 9B is the undisputed champ. Gemma 4 31B is the top dog at s…

> For 16GB laptops, Qwen 3.5 9B is the undisputed champ. you can run qwen 3.6 35BA3B on a 12-16GB vram gpu and ot works pretty well. https://www.youtube.com/watch?v=8F_5pdcD3HY&t=1s even the 27B in some quants can fit. https://www.reddit.com/r/LocalLLaMA/comments/1tkmgwj/qwen27b... qwen IMO is far better for coding, esp agentic coding when combined with something like Pi, it comes probably close enough to Sonnet for…

I want to try a hybrid setup of Gemma 4 E4B with lots of context for general, then Qwen 3.5 9B or larger for coding. Strix Halo set up this weekend, which may enable even larger Qwen models with tons of context.

Re: Gemma 4 12B: A unified, encoder-free multimodal model

#380

Post by ThouYS »

outageroom wrote: post #358

I really like the idea of small models that you can get the most out of. If I weren't a programmer, I wouldn't even know what I would use Opus 4.8 or GPT 5.5 models for.

ironically, coding is the area that small models excel most at. qwen 3.6 27b is an insane agentic model. but Opus can be much more than a programmer: language tutor, knowledgeable "friend", therapist
Post reply on HN