> Rio de Janeiro's city government model... Because... lack of a good open weight LLM is a pressing need high on the municipal priorities list for Rio de Janeiro citizens?
It's the municipal IT company, and the dude that did this is a volunteer.
Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
41–50 of 51 posts
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#42The problem with these is the tool calling. From my experiments qwen agent almost always fails with tool calling and porting the correct config is quite tedious.
Rio3.5 with Qwen compatible tool calling, we need that :)
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#43https://github.com/nex-agi/Nex-N2/issues/4 Seems that they didn't make/train a new novel model, they did a mix of two existing models and then gave it an instruction to say it was 'Rio, trained by Rio AI Labs'
> The model is built via a merge of https://huggingface.co/nex-agi/Nex-N2-Pro and https://huggingface.co/Qwen/Qwen3.5-397B-A17B , proceeded by On-Policy Distillation from a stronger model. We detected an incorrect upload in the previous version, where the base merged version was upload instead of the final distilled model. We are sorry for the confusion and apologize profusely. https://huggingface.co/prefeitura-rio/R…
can you offer a 4-bit quantized version and name it Zé Pequeno, pretty please?
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#44https://github.com/nex-agi/Nex-N2/issues/4 Seems that they didn't make/train a new novel model, they did a mix of two existing models and then gave it an instruction to say it was 'Rio, trained by Rio AI Labs'
> The model is built via a merge of https://huggingface.co/nex-agi/Nex-N2-Pro and https://huggingface.co/Qwen/Qwen3.5-397B-A17B , proceeded by On-Policy Distillation from a stronger model. We detected an incorrect upload in the previous version, where the base merged version was upload instead of the final distilled model. We are sorry for the confusion and apologize profusely. https://huggingface.co/prefeitura-rio/R…
I'd love to see people figuring out how to build models from several smaller ones. We could then train small specialized models and deploy setups more optimized for any given task. Modular LLMs should be a thing.
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#45Earlier quoted context omitted.
> The model is built via a merge of https://huggingface.co/nex-agi/Nex-N2-Pro and https://huggingface.co/Qwen/Qwen3.5-397B-A17B , proceeded by On-Policy Distillation from a stronger model. We detected an incorrect upload in the previous version, where the base merged version was upload instead of the final distilled model. We are sorry for the confusion and apologize profusely. https://huggingface.co/prefeitura-rio/R…
It was a recent edit though. Yesterday snapshot: https://web.archive.org/web/20260613072958/https://huggingfa...
How does that contradict that they uploaded the wrong model?
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#46Earlier quoted context omitted.
> The model is built via a merge of https://huggingface.co/nex-agi/Nex-N2-Pro and https://huggingface.co/Qwen/Qwen3.5-397B-A17B , proceeded by On-Policy Distillation from a stronger model. We detected an incorrect upload in the previous version, where the base merged version was upload instead of the final distilled model. We are sorry for the confusion and apologize profusely. https://huggingface.co/prefeitura-rio/R…
I'd love to see people figuring out how to build models from several smaller ones. We could then train small specialized models and deploy setups more optimized for any given task. Modular LLMs should be a thing.
This is something I've been trying to figure out for a bit, some models are really good at instructions, but their context window is too small, I do wonder if having a cluster of smaller models would be feasible. Been building a custom coding harness so once its nice and polished I might experiment with this more.
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#47Earlier quoted context omitted.
> IMO wherever someone says "the government" we should mentally substitute "we all, collectively". No, we should substitute "unaccountable bureaucrats". The people who enter and leave power from elections are not the source of the daily frustrations people have with government, it's the rest.
how do you think that alleged amorphous mass of unaccountable bureaucrats got their jobs?
Same way anyone else got their job, but that's beside the point. The point is that accountability in government is significantly lower than in private industry, and this is a big source of the problems.
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#48Earlier quoted context omitted.
> IMO wherever someone says "the government" we should mentally substitute "we all, collectively". No, we should substitute "unaccountable bureaucrats". The people who enter and leave power from elections are not the source of the daily frustrations people have with government, it's the rest.
If this is in fact an issue where you life, then you should consider stopping to elect politicians that allow bureaucrats to be unaccountable. Or stop believing politicians who rave on about how bureaucrats are unaccountable while they themselves have the power to shape systems where that would not be the case.
I don't have to believe politicians, I see it with my own eyes whenever I deal with government employees.
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#49Mr Erdoğan launched and initiative yesterday to become the leader in the AI space. As absurd of a claim as his 2023 (hard) landing on the moon.
Re: Rio de Janeiro's city government model Rio3.5 beats Qwen3.7 in recent benchmarks
#50https://github.com/nex-agi/Nex-N2/issues/4 Seems that they didn't make/train a new novel model, they did a mix of two existing models and then gave it an instruction to say it was 'Rio, trained by Rio AI Labs'
To be fair, I still find it to be a great initiative.