gokaygokay/Flux-Watercolor-Strokes-LoRA
Text-to-Image
โข
Updated
โข
132
โข
โข
4
Marc Kovka
GPT007
AI & ML interests
None yet
Recent Activity
liked
a model
19 days ago
gokaygokay/Flux-Watercolor-Strokes-LoRA
Organizations
None yet
GPT007's activity
liked
a
model
19 days ago
reacted to
lewtun's
post with ๐ฅ
19 days ago
Post
6656
We outperform Llama 70B with Llama 3B on hard math by scaling test-time compute ๐ฅ
How? By combining step-wise reward models with tree search algorithms :)
We show that smol models can match or exceed the performance of their much larger siblings when given enough "time to think"
We're open sourcing the full recipe and sharing a detailed blog post.
In our blog post we cover:
๐ Compute-optimal scaling: How we implemented DeepMind's recipe to boost the mathematical capabilities of open models at test-time.
๐ Diverse Verifier Tree Search (DVTS): An unpublished extension we developed to the verifier-guided tree search technique. This simple yet effective method improves diversity and delivers better performance, particularly at large test-time compute budgets.
๐งญ Search and Learn: A lightweight toolkit for implementing search strategies with LLMs and built for speed with vLLM
Here's the links:
- Blog post: HuggingFaceH4/blogpost-scaling-test-time-compute
- Code: https://github.com/huggingface/search-and-learn
Enjoy!
How? By combining step-wise reward models with tree search algorithms :)
We show that smol models can match or exceed the performance of their much larger siblings when given enough "time to think"
We're open sourcing the full recipe and sharing a detailed blog post.
In our blog post we cover:
๐ Compute-optimal scaling: How we implemented DeepMind's recipe to boost the mathematical capabilities of open models at test-time.
๐ Diverse Verifier Tree Search (DVTS): An unpublished extension we developed to the verifier-guided tree search technique. This simple yet effective method improves diversity and delivers better performance, particularly at large test-time compute budgets.
๐งญ Search and Learn: A lightweight toolkit for implementing search strategies with LLMs and built for speed with vLLM
Here's the links:
- Blog post: HuggingFaceH4/blogpost-scaling-test-time-compute
- Code: https://github.com/huggingface/search-and-learn
Enjoy!
- 2 replies
liked
a
model
4 months ago
liked
3
Spaces
4 months ago
liked
a
model
4 months ago
reacted to
m-ric's
post with ๐ฅ
5 months ago
Post
3401
๐๐ผ๐ผ๐ด๐น๐ฒ ๐ฝ๐ฎ๐ฝ๐ฒ๐ฟ : ๐๐ฐ๐ฎ๐น๐ถ๐ป๐ด ๐๐ฝ ๐ถ๐ป๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ฐ๐ฒ ๐ฐ๐ผ๐บ๐ฝ๐๐๐ฒ ๐ฏ๐ฒ๐ฎ๐๐ ๐ญ๐ฐ๐
๐น๐ฎ๐ฟ๐ด๐ฒ๐ฟ ๐บ๐ผ๐ฑ๐ฒ๐น๐ ๐
Remember scaling laws? These are empirical laws that say "the bigger your model, the better it gets". More precisely, "as your compute increases exponentially, loss decreases in a linear fashion". They have wild implications, suggesting that spending 100x more training compute would make you super-LLMs. That's why companies are racing to build the biggest AI superclusters ever, and Meta bought 350k H100 GPUs, which probably cost in the order of $1B.
But think of this : we're building huge reasoning machines, but only ask them to do one pass through the model to get one token of the final answer : i.e., we expend a minimal effort on inference. That's like building a Caterpillar truck and making it run on a lawnmower's motor. ๐๐ต Couldn't we optimize this? ๐ค
๐ก So instead of scaling up on training by training even bigger models on many more trillions of tokens, Google researchers explored this under-explored avenue : scaling up inference compute.
They combine two methods to use more compute : either a reviser that iterated to adapt the model distribution, or generate N different completions (for instance through Beam Search) and select only the best one using an additional verifier model.
They use a Palm-2 model (released in May 23) on the MATH dataset : Palm-2 has the advantage of getting a low performance on MATH, but not zero, so that improvements will be noticeable.
And the results show that for the same fixed amount of inference compute:
๐ฅ a smaller model with more effort on decoding beats a x14 bigger model using naive greedy sampling.
That means that you can divide your training costs by 14 and still get the same perf for the same inference cost!
Take that, scaling laws. Mark Zuckerberg, you're welcome, hope I can get some of these H100s.
Read the paper here ๐ Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters (2408.03314)
Remember scaling laws? These are empirical laws that say "the bigger your model, the better it gets". More precisely, "as your compute increases exponentially, loss decreases in a linear fashion". They have wild implications, suggesting that spending 100x more training compute would make you super-LLMs. That's why companies are racing to build the biggest AI superclusters ever, and Meta bought 350k H100 GPUs, which probably cost in the order of $1B.
But think of this : we're building huge reasoning machines, but only ask them to do one pass through the model to get one token of the final answer : i.e., we expend a minimal effort on inference. That's like building a Caterpillar truck and making it run on a lawnmower's motor. ๐๐ต Couldn't we optimize this? ๐ค
๐ก So instead of scaling up on training by training even bigger models on many more trillions of tokens, Google researchers explored this under-explored avenue : scaling up inference compute.
They combine two methods to use more compute : either a reviser that iterated to adapt the model distribution, or generate N different completions (for instance through Beam Search) and select only the best one using an additional verifier model.
They use a Palm-2 model (released in May 23) on the MATH dataset : Palm-2 has the advantage of getting a low performance on MATH, but not zero, so that improvements will be noticeable.
And the results show that for the same fixed amount of inference compute:
๐ฅ a smaller model with more effort on decoding beats a x14 bigger model using naive greedy sampling.
That means that you can divide your training costs by 14 and still get the same perf for the same inference cost!
Take that, scaling laws. Mark Zuckerberg, you're welcome, hope I can get some of these H100s.
Read the paper here ๐ Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters (2408.03314)
- 1 reply
reacted to
Abhaykoul's
post with ๐ฅ
5 months ago
Post
2969
Introducing HelpingAI2-9B, an emotionally intelligent LLM.
Model Link : https://huggingface.co/OEvortex/HelpingAI2-9B
Demo Link: Abhaykoul/HelpingAI2
This model is part of the innovative HelpingAI series and it stands out for its ability to engage users with emotional understanding.
Key Features:
-----------------
* It gets 95.89 score on EQ Bench greather than all top notch LLMs, reflecting advanced emotional recognition.
* It gives responses in empathetic and supportive manner.
Must try our demo: Abhaykoul/HelpingAI2
Model Link : https://huggingface.co/OEvortex/HelpingAI2-9B
Demo Link: Abhaykoul/HelpingAI2
This model is part of the innovative HelpingAI series and it stands out for its ability to engage users with emotional understanding.
Key Features:
-----------------
* It gets 95.89 score on EQ Bench greather than all top notch LLMs, reflecting advanced emotional recognition.
* It gives responses in empathetic and supportive manner.
Must try our demo: Abhaykoul/HelpingAI2
- 1 reply
liked
2
Spaces
5 months ago
reacted to
alvdansen's
post with ๐ค
5 months ago
Post
6977
Alright Ya'll
I know it's a Saturday, but I decided to release my first Flux Dev Lora.
A retrain of my "Frosting Lane" model and I am sure the styles will just keep improving.
Have fun! Link Below - Thanks again to @ostris for the trainer and Black Forest Labs for the awesome model!
alvdansen/frosting_lane_flux
I know it's a Saturday, but I decided to release my first Flux Dev Lora.
A retrain of my "Frosting Lane" model and I am sure the styles will just keep improving.
Have fun! Link Below - Thanks again to @ostris for the trainer and Black Forest Labs for the awesome model!
alvdansen/frosting_lane_flux
liked
a
Space
5 months ago
upvoted
an
article
5 months ago
Article
Google releases Gemma 2 2B, ShieldGemma and Gemma Scope
โข
59
reacted to
singhsidhukuldeep's
post with ๐
5 months ago
Post
2762
What is the best LLM for RAG systems? ๐ค
In a business setting, it will be the one that gives the best performance at a great price! ๐ผ๐ฐ
And maybe it should be easy to fine-tune, cheap to fine-tune... FREE to fine-tune? ๐ฒโจ
That's @Google Gemini 1.5 Flash! ๐๐
It now supports fine-tuning, and the inference cost is the same as the base model! <coughs LORA adopters> ๐คญ๐ค
So the base model must be expensive? ๐ธ
For the base model, the input price is reduced by 78% to $0.075/1 million tokens and the output price by 71% to $0.3/1 million tokens. ๐๐ต
But is it any good? ๐คทโโ๏ธ
On the LLM Hallucination Index, Gemini 1.5 Flash achieved great context adherence scores of 0.94, 1, and 0.92 across short, medium, and long contexts. ๐๐ฏ
Google has finally given a model that is free to tune and offers an excellent balance between performance and cost. โ๏ธ๐
Happy tuning... ๐ถ๐ง
Gemini 1.5 Flash: https://developers.googleblog.com/en/gemini-15-flash-updates-google-ai-studio-gemini-api/ ๐
LLM Hallucination Index: https://www.rungalileo.io/hallucinationindex ๐
In a business setting, it will be the one that gives the best performance at a great price! ๐ผ๐ฐ
And maybe it should be easy to fine-tune, cheap to fine-tune... FREE to fine-tune? ๐ฒโจ
That's @Google Gemini 1.5 Flash! ๐๐
It now supports fine-tuning, and the inference cost is the same as the base model! <coughs LORA adopters> ๐คญ๐ค
So the base model must be expensive? ๐ธ
For the base model, the input price is reduced by 78% to $0.075/1 million tokens and the output price by 71% to $0.3/1 million tokens. ๐๐ต
But is it any good? ๐คทโโ๏ธ
On the LLM Hallucination Index, Gemini 1.5 Flash achieved great context adherence scores of 0.94, 1, and 0.92 across short, medium, and long contexts. ๐๐ฏ
Google has finally given a model that is free to tune and offers an excellent balance between performance and cost. โ๏ธ๐
Happy tuning... ๐ถ๐ง
Gemini 1.5 Flash: https://developers.googleblog.com/en/gemini-15-flash-updates-google-ai-studio-gemini-api/ ๐
LLM Hallucination Index: https://www.rungalileo.io/hallucinationindex ๐
- 1 reply
reacted to
not-lain's
post with ๐ฅ
5 months ago
Post
6770
๐ฅ New state of the art model for background removal is out
๐ค You can try the model at ZhengPeng7/BiRefNet
๐ model shows impressive results outperforming briaai/RMBG-1.4
๐ you can try out the model in: ZhengPeng7/BiRefNet_demo
๐paper: Bilateral Reference for High-Resolution Dichotomous Image Segmentation (2401.03407)
๐ค You can try the model at ZhengPeng7/BiRefNet
๐ model shows impressive results outperforming briaai/RMBG-1.4
๐ you can try out the model in: ZhengPeng7/BiRefNet_demo
๐paper: Bilateral Reference for High-Resolution Dichotomous Image Segmentation (2401.03407)
- 1 reply
reacted to
Jaward's
post with ๐ฅ
5 months ago
Post
1778
PyTorch implementation of the Self-Compression & Differentiable Quantization Algorithm introduced in โSelf-Compressing Neural Networksโ paper.
The algorithm shows dynamic neural network compression during training - with reduced size of weight, activation tensors and bits required to represent weights.
Itโs basically shrinking the neural network size (weights and activations) as itโs being trained without compromising performance - this helps reduce compute and inference cost.
Code: https://github.com/Jaykef/ai-algorithms
Paper: https://arxiv.org/pdf/2301.13142
The algorithm shows dynamic neural network compression during training - with reduced size of weight, activation tensors and bits required to represent weights.
Itโs basically shrinking the neural network size (weights and activations) as itโs being trained without compromising performance - this helps reduce compute and inference cost.
Code: https://github.com/Jaykef/ai-algorithms
Paper: https://arxiv.org/pdf/2301.13142
reacted to
MonsterMMORPG's
post with โค๏ธ๐๐ฅ
5 months ago
Post
2380
Live Portrait Updated to V5
Animals Live animation added
All of the main repo changes and improvements added to our modified and improve app
Link : https://patreon.com/posts/107609670
Works perfect on Massed Compute, RunPod, free Kaggle account and Windows
1-Click to install with instructions
All tested and verified
Windows tutorial : https://youtu.be/FPtpNrmuwXk
Cloud (RunPod, Massed Compute & free Kaggle account) tutorial : https://youtu.be/wG7oPp01COg
Making XPose / UniPose / ops library compiling working was a challenge on Massed Compute and Kaggle.
Animals Live animation added
All of the main repo changes and improvements added to our modified and improve app
Link : https://patreon.com/posts/107609670
Works perfect on Massed Compute, RunPod, free Kaggle account and Windows
1-Click to install with instructions
All tested and verified
Windows tutorial : https://youtu.be/FPtpNrmuwXk
Cloud (RunPod, Massed Compute & free Kaggle account) tutorial : https://youtu.be/wG7oPp01COg
Making XPose / UniPose / ops library compiling working was a challenge on Massed Compute and Kaggle.