7
Pytorch Day Prez
🔥
Explore PyTorch and Transformer integration in a presentation
Aurélien-Morgan CLAUDON
Aurelien-Morgan
AI & ML interests
None yet
Recent Activity
liked
a Space
2 days ago
Tinkering/Pytorch-day-prez
liked
a model
4 days ago
unsloth/Qwen3-1.7B-Base
published
an
article
7 days ago
retrain-pipelines and the almighty function-caller
Organizations
Aurelien-Morgan's activity
liked
a
Space
2 days ago
liked
a
model
4 days ago
published
an
article
7 days ago
Article
retrain-pipelines and the almighty function-caller
By
•
•
3
posted
an
update
8 days ago
Post
3094
The Almighty function-caller
How would you like to build smart GenAi infrastructure ?
Give extensive tools memory to your edge agentic system,
And optimize the resources it takes to run yet a high-performance set of agents ?
We came up with a novel approach to function-calling at scale for smart companies and corporate-grade use-cases.
Read our full-fledged blog article on this here on Hugging Face :
https://huggingface.co/blog/Aurelien-Morgan/the-almighty-function-caller
How would you like to build smart GenAi infrastructure ?
Give extensive tools memory to your edge agentic system,
And optimize the resources it takes to run yet a high-performance set of agents ?
We came up with a novel approach to function-calling at scale for smart companies and corporate-grade use-cases.
Read our full-fledged blog article on this here on Hugging Face :
https://huggingface.co/blog/Aurelien-Morgan/the-almighty-function-caller
reacted to
danielhanchen's
post with 🔥
9 days ago
Post
5684
🦥 Introducing Unsloth Dynamic v2.0 GGUFs!
Our v2.0 quants set new benchmarks on 5-shot MMLU and KL Divergence, meaning you can now run & fine-tune quantized LLMs while preserving as much accuracy as possible.
Llama 4: unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF
DeepSeek-R1: unsloth/DeepSeek-R1-GGUF-UD
Gemma 3: unsloth/gemma-3-27b-it-GGUF
We made selective layer quantization much smarter. Instead of modifying only a subset of layers, we now dynamically quantize all layers so every layer has a different bit. Now, our dynamic method can be applied to all LLM architectures, not just MoE's.
Blog with Details: https://docs.unsloth.ai/basics/dynamic-v2.0
All our future GGUF uploads will leverage Dynamic 2.0 and our hand curated 300K–1.5M token calibration dataset to improve conversational chat performance.
For accurate benchmarking, we built an evaluation framework to match the reported 5-shot MMLU scores of Llama 4 and Gemma 3. This allowed apples-to-apples comparisons between full-precision vs. Dynamic v2.0, QAT and standard iMatrix quants.
Dynamic v2.0 aims to minimize the performance gap between full-precision models and their quantized counterparts.
Our v2.0 quants set new benchmarks on 5-shot MMLU and KL Divergence, meaning you can now run & fine-tune quantized LLMs while preserving as much accuracy as possible.
Llama 4: unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF
DeepSeek-R1: unsloth/DeepSeek-R1-GGUF-UD
Gemma 3: unsloth/gemma-3-27b-it-GGUF
We made selective layer quantization much smarter. Instead of modifying only a subset of layers, we now dynamically quantize all layers so every layer has a different bit. Now, our dynamic method can be applied to all LLM architectures, not just MoE's.
Blog with Details: https://docs.unsloth.ai/basics/dynamic-v2.0
All our future GGUF uploads will leverage Dynamic 2.0 and our hand curated 300K–1.5M token calibration dataset to improve conversational chat performance.
For accurate benchmarking, we built an evaluation framework to match the reported 5-shot MMLU scores of Llama 4 and Gemma 3. This allowed apples-to-apples comparisons between full-precision vs. Dynamic v2.0, QAT and standard iMatrix quants.
Dynamic v2.0 aims to minimize the performance gap between full-precision models and their quantized counterparts.
posted
an
update
9 days ago
Post
645
retrain-pipelines 0.1.2 finally dropped. It comes with a hot Hugging Face Hub integration. Go check it out. We have 2 articles about it coming up. One already fully written so, be on the lookout !@retrain-pipelines
Also, I'll be volunteering at GOSIM AI Paris 2025. If you're interested in chatting, hmu.
liked
a
Space
10 days ago
upvoted
a
paper
10 days ago
God speed
1
#1 opened 15 days ago
by
Aurelien-Morgan
liked
a
dataset
18 days ago
upvoted
an
article
19 days ago
Article
Cohere on Hugging Face Inference Providers 🔥
•
124
upvoted
an
article
20 days ago
Article
Hugging Face to sell open-source robots thanks to Pollen Robotics acquisition 🤖
•
42
liked
a
Space
20 days ago
liked
a
dataset
21 days ago
Async active filters on HomePage laggy or broken
3
#67 opened 25 days ago
by
Aurelien-Morgan
reacted to
jsulz's
post with 🧠
27 days ago
Post
3106
What does it mean when models share the same bytes?
We've investigated some quants and have seen that a considerable portion of quantizations of the same model share the same bytes and can be deduplicated to save considerable upload time for quantizers on the Hub.
This space where we crack open a repo from @bartowski shows we can get significant dedupe xet-team/quantization-dedup
You can get a sense of why by reading this write-up: https://github.com/bartowski1182/llm-knowledge/blob/main/quantization/quantization.md
But what about finetuned models?
Since going into production the
xet-team
has migrated hundreds of repositories on the Hub to our storage layer, including classic "pre-Hub" open-source models like
FacebookAI/xlm-roberta-large (XLM-R) from
FacebookAI
XLM-R, introduced in 2019, set new benchmarks for multilingual NLP by learning shared representations across 100 languages. It was then fine-tuned on English, Spanish, Dutch, and German, generating language-specific derivations for each - check out the paper here Unsupervised Cross-lingual Representation Learning at Scale (1911.02116)
These finetunes share much of the same architecture and layout as XLM-R with similar training methods and goals. It makes sense that they would share bytes, but it's still fascinating to see.
We put together a similar space to explore these models to see where they overlap - check it out for yourself xet-team/finetune-dedupe
The darker each block in the heatmap, the more the bytes are shared. Clicking on a repos blocks shows all other repos that share blocks.
We've investigated some quants and have seen that a considerable portion of quantizations of the same model share the same bytes and can be deduplicated to save considerable upload time for quantizers on the Hub.
This space where we crack open a repo from @bartowski shows we can get significant dedupe xet-team/quantization-dedup
You can get a sense of why by reading this write-up: https://github.com/bartowski1182/llm-knowledge/blob/main/quantization/quantization.md
But what about finetuned models?
Since going into production the
xet-team
has migrated hundreds of repositories on the Hub to our storage layer, including classic "pre-Hub" open-source models like
FacebookAI/xlm-roberta-large (XLM-R) from
FacebookAI
XLM-R, introduced in 2019, set new benchmarks for multilingual NLP by learning shared representations across 100 languages. It was then fine-tuned on English, Spanish, Dutch, and German, generating language-specific derivations for each - check out the paper here Unsupervised Cross-lingual Representation Learning at Scale (1911.02116)
These finetunes share much of the same architecture and layout as XLM-R with similar training methods and goals. It makes sense that they would share bytes, but it's still fascinating to see.
We put together a similar space to explore these models to see where they overlap - check it out for yourself xet-team/finetune-dedupe
The darker each block in the heatmap, the more the bytes are shared. Clicking on a repos blocks shows all other repos that share blocks.
- 1 reply
upvoted
a
paper
27 days ago
upvoted
an
article
about 1 month ago
Article
The New and Fresh analytics in Inference Endpoints
•
19
[FEEDBACK] Follow
12
25
#14 opened over 1 year ago
by
victor