Pedro Cuenca's picture

Pedro Cuenca

pcuenq

·

AI & ML interests

None yet

Recent Activity

reacted to Molbap's post with ❤️ about 16 hours ago

🚀 New blog: Maintain the unmaintainable – 1M+ Python LOC, 400+ models How do you stop a million-line library built by thousands of contributors from collapsing under its own weight? At 🤗 Transformers, we do it with explicit software-engineering tenets, principles that make the codebase hackable at scale. 🔍 Inside the post: – One Model, One File: readability first — you can still open a modeling file and see the full logic, top to bottom. – Modular Transformers: visible inheritance that cuts maintenance cost by ~15× while keeping models readable. – Config-Driven Performance: FlashAttention, tensor parallelism, and attention scheduling are config-level features, not rewrites. Written with @lysandre,@pcuenq and @yonigozlan, this is a deep dive into how Transformers stays fast, open, and maintainable. Read it here → https://huggingface.co/spaces/transformers-community/Transformers-tenets

reacted to Molbap's post with 🔥 about 16 hours ago

🚀 New blog: Maintain the unmaintainable – 1M+ Python LOC, 400+ models How do you stop a million-line library built by thousands of contributors from collapsing under its own weight? At 🤗 Transformers, we do it with explicit software-engineering tenets, principles that make the codebase hackable at scale. 🔍 Inside the post: – One Model, One File: readability first — you can still open a modeling file and see the full logic, top to bottom. – Modular Transformers: visible inheritance that cuts maintenance cost by ~15× while keeping models readable. – Config-Driven Performance: FlashAttention, tensor parallelism, and attention scheduling are config-level features, not rewrites. Written with @lysandre,@pcuenq and @yonigozlan, this is a deep dive into how Transformers stays fast, open, and maintainable. Read it here → https://huggingface.co/spaces/transformers-community/Transformers-tenets

liked a Space about 16 hours ago

hugging-science/science-release-map

View all activity

Organizations

pcuenq 's models 113

pcuenq/canary-sp-model

Updated 14 days ago

pcuenq/Qwen3-235B-A22B-Instruct-2507-mlx-4Bit

Text Generation • 235B • Updated Aug 14 • 33 • 1

pcuenq/Phi-4-mini-instruct-mlx-4Bit

Text Generation • 0.6B • Updated Aug 14 • 38 • 1

pcuenq/Phi-4-mini-instruct-mlx-8Bit

Text Generation • 1B • Updated Aug 14 • 47

pcuenq/Baichuan-M1-14B-Instruct-tokenizer

Text Generation • Updated Jul 7 • 19

pcuenq/Hunyuan-7B-Instruct-tokenizer

Text Generation • Updated Jul 7 • 15

pcuenq/marigold-coco-segmentation

Updated Jul 4 • 8

pcuenq/t

pcuenq/gated-test

Text Generation • Updated Mar 6

pcuenq/DeepSeek-R1-Distill-Qwen-32B-Q2-6

Text Generation • 4B • Updated Feb 5 • 15 • 1

pcuenq/jh_dreambooth_1000

Text-to-Image • Updated Feb 5 • 47 • 1

pcuenq/gemma-2-2b-it-Q4_K_M

Text Generation • 0.4B • Updated Feb 5 • 12

pcuenq/gpt2-finetuned-cnn-summarization-v2-tokenizer

Text Generation • Updated Jan 31 • 4

pcuenq/smol-lfs

pcuenq/vit_large_patch14_dinov2.lvd142m

Image Feature Extraction • 0.3B • Updated Jan 21 • 40

pcuenq/vit_base_patch32_clip_448.laion2b_ft_in12k_in1k

Image Classification • 0.1B • Updated Jan 20 • 50

pcuenq/tw1

Image Classification • 0.1B • Updated Jan 20 • 7

pcuenq/timm-wrapper-food-lora

pcuenq/vit_base_patch16_224.augreg2_in21k_ft_in1k.lora_ft_food101

pcuenq/distilbert-base-multilingual-cased-tokenizer

Fill-Mask • Updated Jan 16 • 404

pcuenq/convnext_nano.r384_ad_in12k

Image Classification • 0.0B • Updated Jan 16 • 19

pcuenq/gatingtests

pcuenq/gemma-2-2b-it-float32

pcuenq/gemma-2-2b-it-4bit-test

Text Generation • 0.4B • Updated Jan 7 • 5

pcuenq/gemma-2-2b-it-Q4_0-GGUF

Text Generation • 3B • Updated Jan 7 • 20

pcuenq/gemma-2-2b-it-4bit

Text Generation • 0.4B • Updated Jan 7 • 7

pcuenq/tokenizer-lstrip

Updated Dec 12, 2024

pcuenq/misc-examples

Updated Nov 5, 2024

pcuenq/TinyLlama-1.1B-Chat-v1.0-Q8-mlx

0.3B • Updated Oct 17, 2024 • 6

pcuenq/TinyLlama-1.1B-Chat-v1.0-mlx

0.2B • Updated Oct 17, 2024 • 4