Hugging Face's logo
Activity Feed

AI & ML interests

None defined yet.
alvarobarttย 
posted an update 2 months ago
view post
Post
496
Open agents on AWS SageMaker AI with open models from the Hugging Face Hub!

> Deploy an open model from the Hugging Face Hub on SageMaker AI
> Connect the deployed model to Strands Agents
> Add built-in and custom tools for tool calling
> Expose external capabilities through MCP integration
> Bonus: talk to your agent and visualize traces with Gradio

https://alvarobartt.com/agents-on-aws-sagemaker
alvarobarttย 
posted an update 2 months ago
view post
Post
3358
Latest hf-mem release added a breakdown of Mixture-of-Experts (MoE) memory usage!

TL; DR MoEs can be misleading to reason about from active parameters alone, since each token only activates a subset of experts, while the serving setup still needs to account for the full resident memory footprint.

๐Ÿง  hf-mem now splits MoE memory into base model weights, routed experts, and KV cache
๐Ÿ—๏ธ Dense models usually load and use most weights every forward pass, while MoEs load many experts but only route each token to a few of them
โšก Active params isn't the same as memory footprint, especially for sparse architectures
๐Ÿ“ฆ Runtime memory is about what is used per request/token, while loading memory also includes the expert weights that need to be resident
๐Ÿ“š KV cache can still dominate depending on context length, batch size, and concurrency
๐Ÿ”€ Expert Parallelism (EP) helps shard experts across accelerators when expert weights dominate
๐Ÿš€ Data Parallelism (DP) + EP is often a good fit for throughput-oriented MoE serving

Check the repository at https://github.com/alvarobartt/hf-mem
Xiaoqun2023ย 
authored a paper 4 months ago
mishigย 
posted an update 4 months ago
clefourrierย 
authored a paper 4 months ago
alvarobarttย 
posted an update 5 months ago
view post
Post
3762
Learn how to deploy Microsoft Research VibeVoice ASR on Microsoft Azure Foundry with Hugging Face to generate rich audio transcriptions with Who, When, and What! ๐Ÿ’ฅ

> ๐Ÿ•’ 60-minute single-pass processing, no chunking or stitching
> ๐Ÿ‘ค Customized hotwords to guide recognition on domain-specific content
> ๐Ÿ“ Rich transcription: joint ASR + diarization + timestamping in one pass
> ๐ŸŒ 50+ languages with automatic detection and code-switching support
> ๐Ÿค— Deployed on Microsoft Foundry via an OpenAI-compatible Chat Completions API

https://huggingface.co/docs/microsoft-azure/foundry/examples/deploy-vibevoice-asr
mryabย 
submitted a paper to Daily Papers 5 months ago
alvarobarttย 
posted an update 6 months ago
view post
Post
3301
๐Ÿ’ฅ hf-mem v0.4.1 now also estimates KV cache memory requirements for any context length and batch size with the --experimental flag!

uvx hf-mem --model-id ... --experimental will automatically pull the required information from the Hugging Face Hub to include the KV cache estimation, when applicable.

๐Ÿ’ก Alternatively, you can also set the --max-model-len, --batch-size and --kv-cache-dtype arguments (ร  la vLLM) manually if preferred.
  • 1 reply
ยท
alozowskiย 
authored a paper 8 months ago
merveย 
posted an update 9 months ago
view post
Post
13996
deepseek-ai/DeepSeek-OCR is out! ๐Ÿ”ฅ my take โคต๏ธ
> pretty insane it can parse and re-render charts in HTML
> it uses CLIP and SAM features concatenated, so better grounding
> very efficient per vision tokens/performance ratio
> covers 100 languages
  • 5 replies
ยท
jfchiย 
authored 6 papers 10 months ago
merveย 
posted an update 10 months ago
view post
Post
7081
large AI labs open-sourced a ton of models last week ๐Ÿ”ฅ
here's few picks, find even more here merve/sep-16-releases-68d13ea4c547f02f95842f05 ๐Ÿค
> IBM released a new Docling model with 258M params based on Granite (A2.0) ๐Ÿ“ ibm-granite/granite-docling-258M
> Xiaomi released 7B audio LM with base and instruct variants (MIT) XiaomiMiMo/mimo-audio-68cc7202692c27dae881cce0
> DecartAI released Lucy Edit, open Nano Banana ๐ŸŒ (NC) decart-ai/Lucy-Edit-Dev
> OpenGVLab released a family of agentic computer use models (3B/7B/32B) with the dataset ๐Ÿ’ป OpenGVLab/scalecua-68c912cf56f7ff4c8e034003
> Meituan Longcat released thinking version of LongCat-Flash ๐Ÿ’ญ meituan-longcat/LongCat-Flash-Thinking
  • 2 replies
ยท
merveย 
posted an update 10 months ago
view post
Post
3582
IBM just released small swiss army knife for the document models: granite-docling-258M on Hugging Face ๐Ÿ”ฅ

> not only a document converter but also can do document question answering, understand multiple languages ๐Ÿคฏ
> best part: released with Apache 2.0 license ๐Ÿ‘ use it with your commercial projects!
> it supports transformers, vLLM and MLX from the get-go! ๐Ÿค—
> built on SigLIP2 & granite-165M

model: ibm-granite/granite-docling-258M
demo: ibm-granite/granite-docling-258m-demo ๐Ÿ’—
lysandreย 
posted an update 10 months ago
view post
Post
9374
We're kick-starting the process of Transformers v5, with @ArthurZ and @cyrilvallez !

v5 should be significant: we're using it as a milestone for performance optimizations, saner defaults, and a much cleaner code base worthy of 2025.

Fun fact: v4.0.0-rc-1 came out on Nov 19, 2020, nearly five years ago!
  • 6 replies
ยท
merveย 
posted an update 10 months ago
view post
Post
1318
a ton of image/video generation models and LLMs from big labs ๐Ÿ”ฅ

> Meta released facebook/mobilellm-r1-68c4597b104fac45f28f448e, smol LLMs for on-device use ๐Ÿ’ฌ
> Tencent released tencent/SRPO, high res image generation model and tencent/POINTS-Reader, cutting edge OCR ๐Ÿ“
> ByteDance released bytedance-research/HuMo, video generation from any input โฏ๏ธ

find more models, datasets, demos here merve/sep-11-releases-68c7dbfa26bea8cd921fa0ac