Bertrand Charpentier's picture

Bertrand Charpentier

sharpenb

·

https://sharpenb.github.io/

AI & ML interests

Efficient & Reliable ML

Recent Activity

posted an update 1 day ago

How to deploy compressed ML models in your pipeline? We wrote a series of blogs on this topics. Hope that it can be helpful to people: - Standard Model Compression in ML Pipeline: https://www.pruna.ai/blog/standard-model-compression-ml-pipeline - Boost Your Replicate Models with Pruna AI: A Step-by-Step Guide: https://www.pruna.ai/blog/guide-replicate-pruna-ai - Pruna + Triton: A Winning Combination for High-Performance AI Deployments: https://www.pruna.ai/blog/pruna-triton-combination Feel free to join our discord (https://discord.com/invite/rskEr4BZJx) if you have questions ;)

updated a collection 8 days ago

Text-to-text Generation Models (LLMs, Llama, GPT, ...)

View all activity

Organizations

Posts 2

Post

349

How to deploy compressed ML models in your pipeline?

We wrote a series of blogs on this topics. Hope that it can be helpful to people:
- Standard Model Compression in ML Pipeline: https://www.pruna.ai/blog/standard-model-compression-ml-pipeline
- Boost Your Replicate Models with Pruna AI: A Step-by-Step Guide: https://www.pruna.ai/blog/guide-replicate-pruna-ai
- Pruna + Triton: A Winning Combination for High-Performance AI Deployments: https://www.pruna.ai/blog/pruna-triton-combination

Feel free to join our discord (https://discord.com/invite/rskEr4BZJx) if you have questions ;)

Post

1870

We compressed SmolLMs to make 135 variations of them (see https://huggingface.co/PrunaAI?search_models=smolLM) with different quantization configurations with pruna (https://docs.pruna.ai/en/latest/).

We made a blog to summarize our findings (see https://www.pruna.ai/blog/smollm2-smaller-faster) and small LM can be made smaller! :)

models

None public yet

datasets

None public yet