I love how they include a helpful chart that shows this model scores worse than everything else.
Am I looking at the wrong table? It dominates everything on visual interpretation benchmarks.
Edit: specifically ocrbench and VQAv2
Nvidia releases NVLM 1.0 72B open weight model
11–20 of 57 posts
Re: Nvidia releases NVLM 1.0 72B open weight model
#12Reminder that Nvidia is still the only company making any money out of the "AI revolution".
i have yet to hear of anyone actually using AI for something properly
only exception im excited about is the non-main characters from video games, where a lot of the random NPCs, can now actually bring some more fun to the game.
Re: Nvidia releases NVLM 1.0 72B open weight model
#13Reminder that Nvidia is still the only company making any money out of the "AI revolution".
Wrong
Midjourney is profitable. All the acquired startups (i.e. Streamlit or MosaicML) who made millions per employee "made money" for the people who cared.
Re: Nvidia releases NVLM 1.0 72B open weight model
#14Reminder that Nvidia is still the only company making any money out of the "AI revolution".
i have yet to hear of anyone actually using AI for something properly only exception im excited about is the non-main characters from video games, where a lot of the random NPCs, can now actually bring some more fun to the game.
Vision models are a godsent for blind user. I use a vision model to sort my laundry, for instance...
And translation and grammar/spell checking is also at a level which was unthinkable before LLMs hit.
But thats it, really. The "talking machine" aspect of it is more and more uncovered as totally useless.
Re: Nvidia releases NVLM 1.0 72B open weight model
#15It has a non-commercial cc-by-nc-4.0 license, I would guess the only way to use this in production is to use Nvidias data centers to host it? Or are there other ways?
Not a lawyer, not legal advice, but... the legal status quo is that neural network outputs are not copyrightable. They are currently considered not made by humans nor considered a derivative work from the training material / network weights (assuming it's not regurgitating copyrighted material verbatim).
The cc-by-nc-4.0 license applies to the network weights. The only thing non-commercial about the license is that it restricts how you may reproduce the licensed material:
> reproduce and Share the Licensed Material, in whole or in part, for NonCommercial purposes only; and
As long as you are not selling the network weights themselves, nothing in the license prevents you from evaluating the neural network for commercial purposes and selling the outputs. In 'production' you will have to directly download the weights from Nvidia themselves (or another 3rd party which is distributing the network weights non-commercially in good faith) though, you can't share the network weights onto your commercial inference server from another one of your commercial deployment servers. Or at least, it gets more dicy there and may be considered commercial reproduction so better avoid it.
For similar reasons you may 3D print a CC-BY-NC model of a tool and use that tool in your commercial workshop, you may use a CC-BY-NC compiler of a language to compile commercial programs, etc.
Re: Nvidia releases NVLM 1.0 72B open weight model
#16I love how they include a helpful chart that shows this model scores worse than everything else.
I see at a few models (3 models in MMMU) that score lower than Nvidia's. But putting that aside, they at least get points for apparent objectivity. At least they probably aren't fudging numbers.
Re: Nvidia releases NVLM 1.0 72B open weight model
#17Earlier quoted context omitted.
i have yet to hear of anyone actually using AI for something properly only exception im excited about is the non-main characters from video games, where a lot of the random NPCs, can now actually bring some more fun to the game.
Vision models are a godsent for blind user. I use a vision model to sort my laundry, for instance... And translation and grammar/spell checking is also at a level which was unthinkable before LLMs hit. But thats it, really. The "talking machine" aspect of it is more and more uncovered as totally useless.
> I use a vision model to sort my laundry
you built a robot that sorts laundry? Tell us more!
Re: Nvidia releases NVLM 1.0 72B open weight model
#18Earlier quoted context omitted.
Vision models are a godsent for blind user. I use a vision model to sort my laundry, for instance... And translation and grammar/spell checking is also at a level which was unthinkable before LLMs hit. But thats it, really. The "talking machine" aspect of it is more and more uncovered as totally useless.
> I use a vision model to sort my laundry you built a robot that sorts laundry? Tell us more!
No, I never said that. But you already know that. The robot in this case is me holding a smart phone.
Re: Nvidia releases NVLM 1.0 72B open weight model
#19It is a family of multimodal models based on pretrained Qwen2-72B-Instruct LLM and InterViT vision encoder.
There are three variants differentiated by the way the vision tokens are used: decoder-only (like the majority of existing VLM), using cross-attention, and a hybrid. Only the first seems to be on huggingface at the moment.
Also they seem to only train on publically available data, concluding that quality is more important than scale.
Re: Nvidia releases NVLM 1.0 72B open weight model
#20Earlier quoted context omitted.
> I use a vision model to sort my laundry you built a robot that sorts laundry? Tell us more!
No, I never said that. But you already know that. The robot in this case is me holding a smart phone.
Is that faster than just determining by touch what type of garment something is? Or is this about sorting by color?