3 ms·

This article deals with the ginormous models being trained and run in places like meta or openai - but the real AI revolution is a mix of things and we've only

This article deals with the ginormous models being trained and run in places like meta or openai - but the real AI revolution is a mix of things and we've only barely scraped the surface of what we could do with this technology. I say this not because the next ChatGPT will be incredible (which it might be) but because it's actually the open source community which amazes me with what has already been built, and runs locally on consumer hardware. Criticism on the massive cloud models which are expensive to run - is not wrong, for now. But it also used to take an entire room with thousands and thousands of costly vacuum tubes to be able to calculate 2+2. My point is hardware will get better, more efficient and more suitable for AI superficially, just as software will improve and optimize. I bet we'll be able to run bigger and bigger models for lower costs in the long term - but even if running SOA models will remain a premium, the local models still provide incredible value and that's where a lot of the revolution will (and already does) come from.

iLoveOncall 2y ago
I think you're overestimating what the open source community has done. The core value of GenAI (the models) that is currently open sourced has ONLY been contributed by large companies. What the open source *community* has contributed is frontends and implementations of research papers. It is hugely important of course, and not to be ignored, but GenAI simply doesn't exist at all without large companies.
- magic_hamster 2y ago
  Not exactly. The "web uis" are actually products built around these models. There are some truly outstanding examples like comfy, automatic1111 and ollama. But what's been even more impressive to me is how the open source community use these models to create new workflows like texturing in blender. There is just so much to discover yet.
- CuriouslyC 2y ago
  Not true, open source is also doing a lot of fine tunes/post training on Llama3, and there's OpenChat as well, so if Llama didn't exist we'd still have something to work with.
  - iLoveOncall 2y ago
    Both build on top of models made by large companies. Once again, LLMs don't exist without those. And it's obvious given the cost involved in training such models.

iLoveOncall 2y ago

I think you're overestimating what the open source community has done. The core value of GenAI (the models) that is currently open sourced has ONLY been contributed by large companies. What the open source *community* has contributed is frontends and implementations of research papers. It is hugely important of course, and not to be ignored, but GenAI simply doesn't exist at all without large companies.

magic_hamster 2y ago
Not exactly. The "web uis" are actually products built around these models. There are some truly outstanding examples like comfy, automatic1111 and ollama. But what's been even more impressive to me is how the open source community use these models to create new workflows like texturing in blender. There is just so much to discover yet.
CuriouslyC 2y ago
Not true, open source is also doing a lot of fine tunes/post training on Llama3, and there's OpenChat as well, so if Llama didn't exist we'd still have something to work with.
- iLoveOncall 2y ago
  Both build on top of models made by large companies. Once again, LLMs don't exist without those. And it's obvious given the cost involved in training such models.