I'm afraid the performance won't be great. Definitely needs a caching layer.

Great, now please standardize the skills folder and the MCP config.

This is historic. For the first time, an open-weights LLM is right at the top.

We won't be able to run this ourselves, but many providers can.

> For the first time, an open-weights LLM is right at the top.

Hmm, not quite true, I think that honor, for better or worse, goes to OpenAI. When they released GPT2 (or GPT1 for that matter) is was quite literally the SOTA in the ecosystem when it was released.

When OpenAI was actually still a proponent of open AI...

No! because they did not release GPT-2 XL until analogs appeared a year later. So the SOTA model during GPT-2 days was closed weights.

You are correct. I miss the time when OpenAI was open.

Maybe they're animalsonvehiclesmaxxing.

I'm not an expert. Can't we abuse that LLMs don't need to receive audio as a continuous stream without interruptions? Couldn't we just send data and pipe it into the LLM with deduplication (if resending happens)?

  x...y...y[dedup]...z

You’re absolutely correct. A jitter buffer is necessary for a human listener, but a LLM isn’t aware of a time lapse, just like it isn’t aware of the time since your last message in the conversion (unless the chat harness explicitly informs it).

Audio -> ASR - no jitter buffer TTS -> human - jitter buffer

I would like if some of that functionality is extracted in CLI tools. Then every coding agent can use it.

I haven't tried Löve, but I somehow enjoyed reading through the README.md, no AI slop, just a natural writing style with tiny indictors showing the authors' enthusiasm in creating software.

The project predates Github, although the first official release came out about a month prior. So, yeah, not a lot of AI slop 20 years ago.

Useful for tests with LLM interactions.

Oh, you should have a look at Pi then.

https://github.com/badlogic/pi-mono/tree/main/packages/codin...