kosolam

Hey… the code looks good, I’m curious what is your development process? The parts I looked at looked almost like they were written by a human manually.

Not on mobile unfortunately

Why sol is not in the comparison?

Check gpt I think they recently started taking the same route

Impressive compression benchmark. Four times smaller than z standard.

Unfortunately, I just canceled my max account.

Unfortunately, for them.

It doesn’t rival deepseek v4 flash, and of course not deepseek v4 pro. This is my own impression.

Only reason people are interested is it’s free at the moment. I wasn’t impressed by its performance. Once the model gets a price tag it’s usage will be negligible.

Does it mean that subscriptions get more tokens? I’m testing it now for coding instead of claude and it’s very important to understand if I get more due to the price reduction.

Can you share some highlights of what it can do? (I’m just asking the agent to review the pr usually)

What do you mean please give more information.

״I am not sure this laptop has a fan anymore. Except me״ huh nice one

Lol

But codex is opensource, no?

Looks like here lua vm is used as the basis for the project itself not as the language u write your builds with

Doing similar stuff. I curse a lot. The processes do transfer mostly between different models which a bit validates the approach.

I never looked back. Full control, and it’s really basic stuff you can have in an hour or so. And then you stop fighting with the external dependency and can move fast with your ideas.

Ha I’m doing the same. 2 months in or so, lost track of time a bit. Started with running all the usual suspects pi, claude code, codex, goose, but mostly migrated to own agent (ha!) for greater level of control. Also doing it in python for C lang as first target (ha!). Anyhow post a link if you’d release yours I’d be interested to take a look at adjacent work.

They dont give the exact size hmm

Hey Dave, I’d love to add your new model in the harness I’m going to opensource very soonish. Going to publish benchmarks on real world tasks.

Yeah, pretty much aligns with my own observations… I think the rage attacks began when Opus 4.7 was released more or less kah kha

But wait, I’m a user of tmux or zellij which for me was basically another tmux and I went back to tmux eventually. I’m not looking to develop another tmux or something. Is this project more of a library to create tmux like apps?

Hey, I automate tmux all the time with send keys and capture.. how is your project improves vs tmux or zellij?

Hey, this is most probably related to the chat template or the reasoning parser or the tool call parser or also things like kv cache quantization and possibly other params that affect results like the regular top k top p and all of that, the backend often sets its own defaults or the lack of them. It’s best to have all these under control if possible. I wonder regarding this project have you been testing it on real world projects? I’m working on an agentic loop as well also using a local model.

Nice. Very clean api.

For anyone that ever reads this, the issue was associated with the zoom level, up to a certain zoom level it behaves ok and wraps the text.

Worked fine for many years for me. Not great, but usable.

I mean in portrait. In landscape it’s fine.