owensong/Inflect-Nano-v1
Text-to-Speech • Updated • 119
appvoid
appvoid
AI & ML interests
Working on symbolic models
Recent Activity
liked a model 1 day ago
owensong/Inflect-Nano-v1 repliedto their post 4 days ago
Everyone is talking about Anthropic’s Fable 5.
Bigger models. Longer tasks. More power.
That’s exciting, but if the future of AI only belongs to people with massive compute, then it is not really for everyone.
That’s why the Hugging Face smol models event matters.
Small models are not just tiny versions of big models.
They are the models that can run cheaper, faster, closer to the user, and eventually on normal machines.
That is the lane I’m building in.
I’m working on a rewrite model series for messy real-world input:
voice transcripts, rambling thoughts, half-correct instructions, and all the “no wait, actually…” moments where the intent is there, but buried.
A giant model can clean that up.
Cool.
I want small open models to get better at it too.
Current goal:
$54.43/month
Colab Pro+ + fees
11% funded
That lets me iterate faster with better datasets, techniques, and compute. Yep.
If you care about useful small models and open experimentation, even $1 helps:
https://ko-fi.com/appvoid
If you are curious about my work, you can check my latest palmer version. That should give you an idea of my recent work.
🤏 smol for the win! 💪
https://huggingface.co/appvoid/palmer-005-nano posted an update 4 days ago
Everyone is talking about Anthropic’s Fable 5.
Bigger models. Longer tasks. More power.
That’s exciting, but if the future of AI only belongs to people with massive compute, then it is not really for everyone.
That’s why the Hugging Face smol models event matters.
Small models are not just tiny versions of big models.
They are the models that can run cheaper, faster, closer to the user, and eventually on normal machines.
That is the lane I’m building in.
I’m working on a rewrite model series for messy real-world input:
voice transcripts, rambling thoughts, half-correct instructions, and all the “no wait, actually…” moments where the intent is there, but buried.
A giant model can clean that up.
Cool.
I want small open models to get better at it too.
Current goal:
$54.43/month
Colab Pro+ + fees
11% funded
That lets me iterate faster with better datasets, techniques, and compute. Yep.
If you care about useful small models and open experimentation, even $1 helps:
https://ko-fi.com/appvoid
If you are curious about my work, you can check my latest palmer version. That should give you an idea of my recent work.
🤏 smol for the win! 💪
https://huggingface.co/appvoid/palmer-005-nanoOrganizations
liked a
model 1 day ago
replied to their post 4 days ago
Also, you can help by sharing ideas directly to me. Bonus points if byte-level related architectures.
posted an update 4 days ago
Post
54
Everyone is talking about Anthropic’s Fable 5.
Bigger models. Longer tasks. More power.
That’s exciting, but if the future of AI only belongs to people with massive compute, then it is not really for everyone.
That’s why the Hugging Face smol models event matters.
Small models are not just tiny versions of big models.
They are the models that can run cheaper, faster, closer to the user, and eventually on normal machines.
That is the lane I’m building in.
I’m working on a rewrite model series for messy real-world input:
voice transcripts, rambling thoughts, half-correct instructions, and all the “no wait, actually…” moments where the intent is there, but buried.
A giant model can clean that up.
Cool.
I want small open models to get better at it too.
Current goal:
$54.43/month
Colab Pro+ + fees
11% funded
That lets me iterate faster with better datasets, techniques, and compute. Yep.
If you care about useful small models and open experimentation, even $1 helps:
https://ko-fi.com/appvoid
If you are curious about my work, you can check my latest palmer version. That should give you an idea of my recent work.
🤏 smol for the win! 💪
appvoid/palmer-005-nano
Bigger models. Longer tasks. More power.
That’s exciting, but if the future of AI only belongs to people with massive compute, then it is not really for everyone.
That’s why the Hugging Face smol models event matters.
Small models are not just tiny versions of big models.
They are the models that can run cheaper, faster, closer to the user, and eventually on normal machines.
That is the lane I’m building in.
I’m working on a rewrite model series for messy real-world input:
voice transcripts, rambling thoughts, half-correct instructions, and all the “no wait, actually…” moments where the intent is there, but buried.
A giant model can clean that up.
Cool.
I want small open models to get better at it too.
Current goal:
$54.43/month
Colab Pro+ + fees
11% funded
That lets me iterate faster with better datasets, techniques, and compute. Yep.
If you care about useful small models and open experimentation, even $1 helps:
https://ko-fi.com/appvoid
If you are curious about my work, you can check my latest palmer version. That should give you an idea of my recent work.
🤏 smol for the win! 💪
appvoid/palmer-005-nano
- 1 reply
posted an update 14 days ago
Post
150
yikes! i missed the small model hackathon i guess i'll have to make sota for people to notice
posted an update 15 days ago
Post
97
model | epoch | step | val_loss |
c-gpt-1| 5 | 41365 | 1.0175 |
If you don't understand what you see, this is just one the coolest proof of concepts I've ever made. I just trained a char-level, super small model (~80 million parameters) on millions of high-quality examples on a curated dataset and is slowly getting davinci-003 vibes
c-gpt-1| 5 | 41365 | 1.0175 |
If you don't understand what you see, this is just one the coolest proof of concepts I've ever made. I just trained a char-level, super small model (~80 million parameters) on millions of high-quality examples on a curated dataset and is slowly getting davinci-003 vibes
liked a
model 20 days ago
posted an update 23 days ago
Post
177
As an advocate for small language models I just want to say. It might not actually be the end for small models. We are just getting started! Now that we have super good models we can find creative ways to replicate the behavior at small scale!
I'll show you in a few weeks what a small model is capable of, you will surprised.
I'll show you in a few weeks what a small model is capable of, you will surprised.
- 1 reply
Great!
updated a
collection 26 days ago
updated a
dataset about 1 month ago
updated a
model about 1 month ago
posted an update about 1 month ago
Post
199
Cheers for a year of sota AI on cpus 🥂 people actually liked my last model, here's another sota for you. This one should feel way different in terms of quality.
appvoid/palmer-005-core
appvoid/palmer-005-core
Real Contribution Here
🚀 1
1
#1 opened about 1 month ago
by 
BingoBird
replied to their post about 1 month ago
Next release will be above most 1b models while being 1/3 the size
replied to their post about 1 month ago
updated a
collection about 1 month ago
updated a
model about 1 month ago
posted an update about 1 month ago
Post
140
new model tomorrow, less than 100m params, 1b level on text editing tasks
🤖 palmer is coming back...
🤖 palmer is coming back...
- 1 reply
updated a
collection about 1 month ago