Running Gemma 4 26B at 5 tokens/sec on a 13-year-old Xeon with no GPU

I have a prediction. By the mid of 2027, we will have >200B MoE models running on basic consumer hardware.

I am running Qwen3.6-35B-A3B locally on my 16GB mac with 7-9 tokens/second. Link - https://github.com/deepanwadhwa/samosa-chat

This is a GPT4 level model running locally with a decent speed on a 16gb ram macbook air.

Some ppl don't like to hear it. But I would assume that token costs when using an inference provider are cheaper than electricity of using locally.

If we just take into account output token generation for simplicity. With 5tps u get 18k tokens an hour. That would costs around 0.005USD from an inference provider.

I estimate that the server consumes probably around 500W during inference.

In Germany where 1kwh cost around 0.3USD, 18k tokens inferred locally would therefore cost 0.15USD which is 30x the costs of using an inference provider.

But for ppl who worry about their data, running locally might still be good. However, they should be aware, that it is much less efficient than using an inference provider.

The efficiency gap will also significantly increase as new GPUs will make inference much more efficient.

EDIT: I first thought it'd be 180k token, but thanks to someone mentioning in the comments, it is 18k. I guess with that, it will be tough unless u got electricity almost for free. Also, the inference providers are probably still using H200/H100 for those small models. Once they use GB300 or next year the new Ruby GPUs, inference will be cheaper by a factor of 30. By then, running local models will mostly be about privacy.

Here's my report running several different models on a dual Xeon with 256 GB of DDR4 and no GPU.

https://gist.github.com/hparadiz/f3596d00a62d8ebb2dadcc46ee5...

That's quite slow I'm getting 8-12 t/s on a 13 year old CPU. (Speed varies by context size and other settings who knows)

https://news.ycombinator.com/item?id=48354801

A dual Xeon of this era is probably pulling 300W or more when loaded.

At national average electricity prices, that’s $1.35 per day. More during the summer if you have to cool the space.

If you run it 24/7 and ignore prompt processing time (not a good assumption at all) it would get around 400,000 tokens in a day.

That’s about $0.30 per million output tokens.

Coincidentally, that’s the same price for this model on OpenRouter right now, but OpenRouter token gen will be 8X faster.

There are a lot of good reasons to experiment with running LLMs locally, like if you don’t want any data leaving your house.

Don’t think that you’re going to come out ahead monetarily. I say this as someone with a lot more money invested in local inference hardware at home. It’s fun, but it’s not a way to save money.

Author here, it looks like my original comment was flagged for some reason. The fix is open upstream as PR #2138 (https://github.com/ikawrakow/ik_llama.cpp/pull/2138)

He's shown me his set up in his basement. It's sick! Talk about your 3d printer next!

I love my little dual core X99 board with Xeon E5 2673 V3. It's not power efficient, but I just leave it in my basement for local Jupyter Notebook stuff. Much faster than everything cloud-based for a reasonably price at my scale.

How much RAM did this need?

I run the same setup Gemma 4 26B on a 2013 Mac Pro (dual graphics cards but they're useless for this). I also get about 5 t/s. It's perfectly serviceable for some tasks!

I was inspired by that post also, got a Qwen Coder 1.5B up to 27tok/s prompt eval and 13tok/s decode on an e5-2650v2 inside a GNOME box

Good job. Thanks for sharing. I have a similar NAS server with 2 Intel(R) Xeon(R) CPU E5-2699 CPUs. I will test as well.

The transformer architecture is fundamentally unsuitable for local inference, while being efficient at scale. It's a fun experiment to try, but that's about it.

Is it just me or does this post not mention how much RAM they had? I would love to know - I have a dual-Xeon 1U screamer with 96GB of DDR4 RDIMM just sitting around...

FWIW I'm getting a hardware max of 20 tok/s (approx topping out the GPU's compute) on my custom local diffusiongemma port running on an M3.

Truly amazing. This gives a peek into the future for what's possible.

Need to run this on my Xeons with AMX

Related:

A 10 year old Xeon is all you need

https://news.ycombinator.com/item?id=48353348

To me context means everything. Tokens per second is a great metric but in the real world context window is the deal breaker when a real use case is on the table.

...and how many servings can this do?

Author here. The short version: a viral post ran Gemma 4 on a 2016 Xeon; my Xeons are 2013, and the fork it used assumes AVX2, which Ivy Bridge doesn't have. The build failure was easy. The fun bug was the silent one: two MoE graph ops with no dispatch case on non-AVX2 builds, so every expert FFN output was uninitialized memory. Deterministic, NaN-free, fluent-looking multilingual gibberish.

The fix is open upstream as PR #2138 (https://github.com/ikawrakow/ik_llama.cpp/pull/2138), awaiting review. Fair warning on the AI angle: the patch was written by Claude at my direction. The post is explicit about which parts were me and which weren't. Happy to answer questions about either the bug or the workflow.

[dead]

[dead]

Apologies for asking here but literally nobody knows:

Android studio connected to a local model disconnects automatically after 10 minutes. How set this limit to 12 hours or remove it completely?

I could run my LM studio model all night... but I cant, since Android studio times out after a hard limit of 10M.

This is not related to number of tokens.

I tried Googling, searching for settings in Android studio, even created a stackoverflow post - but zero information. Jetbrains mentions "remote agent timeout mechanism" - but after changing it, nothing happens.

Unfortunately, the post comes off as AI-written. Why not just write your own posts?