Soumith Chintala
@soumithchintala.bsky.social
Cofounded and lead PyTorch at Meta. Also dabble in robotics at NYU.
AI is delicious when it is accessible and open-source.
http://soumith.ch
AI is delicious when it is accessible and open-source.
http://soumith.ch
i added an example here: bsky.app/profile/soum...
i'll give a representative but not exact example:
change the color of X's shirt from blue to red: the generations often change the entire shirt style itself -- they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
change the color of X's shirt from blue to red: the generations often change the entire shirt style itself -- they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
January 1, 2025 at 7:57 PM
i added an example here: bsky.app/profile/soum...
what I'm finding is that, the models want to be more of an artist than a replacement for photoshop -- which is fine, but I want to be the artist here, and want the tool to be more of a "magically easier photoshop where I ask it what to do in detail, and it does that -- not more not less"
January 1, 2025 at 7:56 PM
what I'm finding is that, the models want to be more of an artist than a replacement for photoshop -- which is fine, but I want to be the artist here, and want the tool to be more of a "magically easier photoshop where I ask it what to do in detail, and it does that -- not more not less"
i'll give a representative but not exact example:
change the color of X's shirt from blue to red: the generations often change the entire shirt style itself -- they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
change the color of X's shirt from blue to red: the generations often change the entire shirt style itself -- they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
January 1, 2025 at 7:56 PM
i'll give a representative but not exact example:
change the color of X's shirt from blue to red: the generations often change the entire shirt style itself -- they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
change the color of X's shirt from blue to red: the generations often change the entire shirt style itself -- they don't respect how much and what I'm trying to change, and dont try to preserve details I ask to preserve
3. They've also made it easy to load MJCF and other common specs used in robotics. They've also made visualization work out of the box (they hacked up a hybrid of pyrender, pyglet and LuisaRender with a ton of their own patches).
December 20, 2024 at 9:03 PM
3. They've also made it easy to load MJCF and other common specs used in robotics. They've also made visualization work out of the box (they hacked up a hybrid of pyrender, pyglet and LuisaRender with a ton of their own patches).
2. The APIs are reasonably simple and well-designed, and they did take out the cross-platform pain in many ways -- CPU, CUDA, Metal etc. are all supported across Linux, OSX, Windows -- thanks to Taichi (and to a small part PyTorch).
December 20, 2024 at 9:03 PM
2. The APIs are reasonably simple and well-designed, and they did take out the cross-platform pain in many ways -- CPU, CUDA, Metal etc. are all supported across Linux, OSX, Windows -- thanks to Taichi (and to a small part PyTorch).
1. It's nice that the internals are written with Taichi, so all the sim code is written in python, more accessible and easy-to-read than retrofitting physics on top of a Tensor compiler (like mujoco did with MJX) and possibly faster because Taichi is a more suited DSL / compiler.
December 20, 2024 at 9:03 PM
1. It's nice that the internals are written with Taichi, so all the sim code is written in python, more accessible and easy-to-read than retrofitting physics on top of a Tensor compiler (like mujoco did with MJX) and possibly faster because Taichi is a more suited DSL / compiler.
The whole GenAI/LLM/VLM stuff seems to be unreleased or "aspirational".
My favorite aspects:
My favorite aspects:
December 20, 2024 at 9:03 PM
The whole GenAI/LLM/VLM stuff seems to be unreleased or "aspirational".
My favorite aspects:
My favorite aspects:
It's basically like Mujoco but with more advanced materials/rendering/solvers, written all in Python thanks to being powered by Taichi, which makes it much more accessible.
I like it a lot. It's very accessible.
They went too far with marketing, but willing to ignore it for now.
I like it a lot. It's very accessible.
They went too far with marketing, but willing to ignore it for now.
December 20, 2024 at 9:03 PM
It's basically like Mujoco but with more advanced materials/rendering/solvers, written all in Python thanks to being powered by Taichi, which makes it much more accessible.
I like it a lot. It's very accessible.
They went too far with marketing, but willing to ignore it for now.
I like it a lot. It's very accessible.
They went too far with marketing, but willing to ignore it for now.
also, congrats OpenAI on O3, and thank you for rapidly making progress on intelligence.
December 20, 2024 at 8:59 PM
also, congrats OpenAI on O3, and thank you for rapidly making progress on intelligence.
Models are dumb as rock without the right context -- pretrained context doesn't help with day-to-day or specialized things.
Private ecosystems and company bureaucracies means you have to feed the models your own context for the next X years....unless computer-use gets ready.
Cant wait for it!
Private ecosystems and company bureaucracies means you have to feed the models your own context for the next X years....unless computer-use gets ready.
Cant wait for it!
December 20, 2024 at 8:59 PM
Models are dumb as rock without the right context -- pretrained context doesn't help with day-to-day or specialized things.
Private ecosystems and company bureaucracies means you have to feed the models your own context for the next X years....unless computer-use gets ready.
Cant wait for it!
Private ecosystems and company bureaucracies means you have to feed the models your own context for the next X years....unless computer-use gets ready.
Cant wait for it!
so much detail, it's incredible that you've gotten this deep....twice ☺️!!!
November 19, 2024 at 12:20 PM
so much detail, it's incredible that you've gotten this deep....twice ☺️!!!