I ran these in LM Studio and got unrecognizable pelicans out of the 2B and 4B models and an outstanding pelican out of the 26b-a4b model - I think the best I've seen from a model that runs on my laptop. https://simonwillison.net/2026/Apr/2/gemma-4/ The gemma-4-31b model is completely broken for me - it just spits out "---\n" no matter what prompt I feed it. I got a pelican out of it via the AI Studio API hosted model…
Do you have a single gallery page where we can see all the pelicans together. I'm thinking something similar to
Hi all! I work on the Gemma team, one of many as this one was a bigger effort given it was a mainline release. Happy to answer whatever questions I can
Do any of you use this as a replacement for Claude Code? For example, you might use it with openclaw. I have a 24 GB integrated RAM Mac Mini M4 I currently run Claude Code on, do you think I can replace it with OpenClaw and one of these models?
Following as I also don’t love the idea of double paying anthropic for my usage plan and API credits to feed my pet lobster.
I'm curious about the multimodal capabilities on the E2B and E4B and how fast is it. In ChatGPT right now, you can have a audio and video feed for the AI, and then the AI can respond in real-time. Now I wonder if the E2B or the E4B is capable enough for this and fast enough to be run on an iPhone. Basically replicating that experience, but all the computations (STT, LLM, and TTS) are done locally on the phone. I just…
yeah, it appears to support audio and image input.. and runs on mobile devices with 256K context window!
Hi all! I work on the Gemma team, one of many as this one was a bigger effort given it was a mainline release. Happy to answer whatever questions I can
What’s it like to work on the frontier of AI model creation? What do you do in your typical day?
I’ve been really enjoying using frontier LLMs in my work, but really have no idea what goes into making one.
Hi all! I work on the Gemma team, one of many as this one was a bigger effort given it was a mainline release. Happy to answer whatever questions I can
Is distillation or synthetic data used during pre-training? If yes how much?
Public benchmarks can be trivially faked. Lmarena is a bit harder to fake and is human-evaluated. I agree it's misleading for them to hyper-focus on one metric, but public benchmarks are far from the only thing that matters. I place more weight on Lmarena scores and private benchmarks.
Lm arena is so easy to game that it's ceased to be a relevant metric over a year ago. People are not usable validators beyond "yeah that looks good to me", nobody checks if the facts are correct or not.
Alibaba maintains its own separate version of lm-arena where the prompts are fixed and you simply judge the outputs
Google might not have the best coding models (yet) but they seem to have the most intelligent and knowledgeable models of all especially Gemini 3.1 Pro is something. One more thing about Google is that they have everything that others do not: 1. Huge data, audio, video, geospatial 2. Tons of expertise. Attention all you need was born there. 3. Libraries that they wrote. 4. Their own data centers and cloud. 4. Most of…
Tiny AI labs with a fraction of Google's resources still turn out amazing open weights. But besides the logistics, the other aspect is can I use it? Gemini (and some other models) have a habit of dropping conversations altogether if it's "uncomfortable" with your question. Recently I was just asking it about financial implications of the war. It decided my ideas were so crazy that I must be upset, and refused to tell me anything else about finance in that chat. Whereas other models (not abliterated, just normal models) gave me information without argument, moralizing, or gaslighting. I think most people are gonna prefer the non-nerfed models, even if they aren't SOTA, because nobody wants to have an argument with their computer.
Still pretty unusable on Raspberry Pi 5, 16gb despite saying its built for it, from the E4B model total duration: 12m41.34930419s load duration: 549.504864ms prompt eval count: 25 token(s) prompt eval duration: 309.002014ms prompt eval rate: 80.91 tokens/s eval count: 2174 token(s) eval duration: 12m36.577002621s eval rate: 2.87 tokens/s Prompt: whats a great chicken breast recipe for dinner tonight?
On my MBP M4 Pro 48gb same model/question while multitasking with Figma, email etc:
Gemma 3 E4E runs very quick on my Samsung S26, so I am looking forward to trying Gemma 4! It is fantastic to have local alternatives to frontier models in an offline manner.
What's the easiest way to install these on an Android phone/Samsung?