You can do coding and agentic fine. For coding I use qwen3.6:35b-mlx and agentic granite4.1:3b works fine.

These are the models I use.

- granite4.1:3b

- granite4.1:30b

- gpt-oss:20b

- gpt-oss:120b (less so now)

- mistral-small3.2

- qwen3.6:35b-mlx

There will always be use cases that don't sit on your laptop, but most of what can be done can be done locally, it just requires a good framework to sit on it.

Even faster with the MLX builds.

Then when I need more heavy lifting I fire up a larger model.

IMHO the issue isn't the models. I've had OpenClaw give the same results as Claude using open models locally. Slower but does the job. Something that can do optimal model switching is what's needed.

It's like running a sports car and then complaining it burns through petrol too fast.

The truth is the model while impressive is not needed for much of what people need.

Local models can do the work and just offload heavy lifting to the cloud models.

https://huggingface.co/ibm-granite

I think if you were genuinely interested, you could have found this yourself.

The granite site covers everything you keep asking for. Granite is made using lm-engine and the details are there.

Without the weights you are not going to be able to build to the same level of accuracy without some serious work.

> the complete recipes to make the models.

You mean the weights which most companies don't release. Again you can find from that link.

https://github.com/ibm-granite

If you check HF you will see its Apache2 and the datasets were also permissive.

It's one of the few models on the market where the creator indemnifies it against copyright claims.

https://research.ibm.com/blog/granite-ethical-ai

Standard rule is you never let your developers at the production instance. So I can't see why an LLM would get a break.

Even with the weights the extra context allows it to move to the correct space.

Much the same as humans there are terms that are meaningless without knowing the context.

You can certainly prove that mathematically, and giving that proof to an LLM it will give you the correct answer. Which is a prompting technique btw to improve accuracy of an LLMs results.

Although the more recent LLMs have built in approaches to answer this question.

Gemma4 for example:

-----

The user is asking for the number of 'r's in the word "strawberry".

* S

* T

* *R* (1)

* A

* W

* B

* E

* *R* (2)

* *R* (3)

* Y

There are 3 'r's.

-----

GPT-OSS-120b example.

-----

The user asks: "How many r's in strawberry." This is a straightforward counting problem: count the letter 'r' in the word "strawberry". The word "strawberry" contains letters: s t r a w b e r r y. Count r's: there are three 'r's? Let's check: s(1) t(2) r(3) a(4) w(5) b(6) e(7) r(8) r(9) y(10). Actually the word is "strawberry". Let's write out: s, t, r, a, w, b, e, r, r, y. So there are three r's: at positions 3, 8, 9. Yes, three r's. So answer: 3.

But we may also note that "r" appears three times.

Thus answer: 3.

We can provide a short answer.

Thus final: There are three r's in "strawberry".

----

Anecdotal, but I've noticed while this is true it also adds the danger of knowing when to stop.

Early on I would take forever trying to get something exactly to whats in my head. Which meant I would spend too much time in one sitting then if I had previously built it by hand.

Now I try to time box with the mindset "good enough".

It was only temporarily banned. It's currently still on the App store since 2017.