π€ Serve any model with Inference Endpoints + Custom Handlers
β’
3
Alvaro Bartolome
alvarobartt
AI & ML interests
βοΈ cloud machine learning @huggingface and open source passionate
Recent Activity
upvoted
a
paper
5 days ago
Spectrum: Targeted Training on Signal to Noise Ratio
liked
a model
6 days ago
Qwen/Qwen-VL-Chat
updated
a model
7 days ago
merve/paligemma_vqav2
Articles
Organizations
alvarobartt's activity
reacted to
AdinaY's
post with ππ₯
26 days ago
Post
1586
π The wave of reasoning models from the Chinese community has arrived!
π Marco-o1 by AIDC, Alibaba
π AIDC-AI/Marco-o1
β¨ QwQ by Qwen, Alibaba
π Qwen/qwq-674762b79b75eac01735070a
π Skywork-o1 by Kunlun Tech
π Skywork/skywork-o1-open-67453df58e12f6c3934738d0
π₯ Xkev/Llama-3.2V-11B-cot by PKU Yuan group
π Xkev/Llama-3.2V-11B-cot
π‘ DeepSeek-R1-Lite-Preview by DeepSeek AI
π https://chat.deepseek.com/
π InternThinker Preview by Shanghai AI Lab
π https://sso.openxlab.org.cn/login?redirect=https://internlm-chat.intern-ai.org.cn/&clientId=ebmrvod6yo0nlzaek1yp
π k0-math by Moonshot AI
π https://kimi.moonshot.cn/ ( coming soon! )
Who's next? π
zh-ai-community/reasoning-models-67409fb3aa1ed78f10087cd7
π Marco-o1 by AIDC, Alibaba
π AIDC-AI/Marco-o1
β¨ QwQ by Qwen, Alibaba
π Qwen/qwq-674762b79b75eac01735070a
π Skywork-o1 by Kunlun Tech
π Skywork/skywork-o1-open-67453df58e12f6c3934738d0
π₯ Xkev/Llama-3.2V-11B-cot by PKU Yuan group
π Xkev/Llama-3.2V-11B-cot
π‘ DeepSeek-R1-Lite-Preview by DeepSeek AI
π https://chat.deepseek.com/
π InternThinker Preview by Shanghai AI Lab
π https://sso.openxlab.org.cn/login?redirect=https://internlm-chat.intern-ai.org.cn/&clientId=ebmrvod6yo0nlzaek1yp
π k0-math by Moonshot AI
π https://kimi.moonshot.cn/ ( coming soon! )
Who's next? π
zh-ai-community/reasoning-models-67409fb3aa1ed78f10087cd7
reacted to
andito's
post with π₯π
27 days ago
Post
1784
SmolVLM speeding locally on a laptop thanks to mlx-vlm and
@Gradio ! Try it with two lines:
pip install git+https://github.com/andimarafioti/mlx-vlm.git@stream-generate-fix
python -m mlx_vlm.chat_ui --model mlx-community/SmolVLM-Instruct-8bit
Gotta love the MLX community! Big thanks to @pcuenq and @prince_canuma !
@Gradio ! Try it with two lines:
pip install git+https://github.com/andimarafioti/mlx-vlm.git@stream-generate-fix
python -m mlx_vlm.chat_ui --model mlx-community/SmolVLM-Instruct-8bit
Gotta love the MLX community! Big thanks to @pcuenq and @prince_canuma !
reacted to
danielhanchen's
post with π₯
about 1 month ago
Post
1347
Vision finetuning is in π¦₯Unsloth! You can now finetune Llama 3.2, Qwen2 VL, Pixtral and all Llava variants up to 2x faster and with up to 70% less VRAM usage! Colab to finetune Llama 3.2: https://colab.research.google.com/drive/1j0N4XTY1zXXy7mPAhOC1_gMYZ2F2EBlk?usp=sharing
- 1 reply
reacted to
clem's
post with π₯
2 months ago
Post
4436
This is no Woodstock AI but will be fun nonetheless haha. Iβll be hosting a live workshop with team members next week about the Enterprise Hugging Face hub.
1,000 spots available first-come first serve with some surprises during the stream!
You can register and add to your calendar here: https://streamyard.com/watch/JS2jHsUP3NDM
1,000 spots available first-come first serve with some surprises during the stream!
You can register and add to your calendar here: https://streamyard.com/watch/JS2jHsUP3NDM
- 4 replies
posted
an
update
4 months ago
Post
2848
π€ Serving Meta Llama 3.1 405B on Google Cloud is now possible via the Hugging Face Deep Learning Containers (DLCs) for Text Generation Inference (TGI)
In this post, we showcase how to deploy https://huggingface.co./meta-llama/Meta-Llama-3.1-405B-Instruct-FP8 on an A3 instance with 8 x H100 GPUs on Vertex AI
Thanks to the Hugging Face DLCs for TGI and Google Cloud Vertex AI, deploying a high-performance text generation container for serving Large Language Models (LLMs) has never been easier. And weβre not going to stop here β stay tuned as we enable more experiences to build AI with open models on Google Cloud!
Read the full post at https://huggingface.co./blog/llama31-on-vertex-ai
In this post, we showcase how to deploy https://huggingface.co./meta-llama/Meta-Llama-3.1-405B-Instruct-FP8 on an A3 instance with 8 x H100 GPUs on Vertex AI
Thanks to the Hugging Face DLCs for TGI and Google Cloud Vertex AI, deploying a high-performance text generation container for serving Large Language Models (LLMs) has never been easier. And weβre not going to stop here β stay tuned as we enable more experiences to build AI with open models on Google Cloud!
Read the full post at https://huggingface.co./blog/llama31-on-vertex-ai
reacted to
merve's
post with π₯
5 months ago
Post
4973
Idefics3-Llama is out! π₯π₯
Model: HuggingFaceM4/Idefics3-8B-Llama3
Demo: HuggingFaceM4/idefics3
It's a multimodal model based on Llama 3.1 that accepts an arbitrary number of interleaved images with text with a huge context window (10k tokens!) β¨
Supported by Hugging Face transformers π€
Model: HuggingFaceM4/Idefics3-8B-Llama3
Demo: HuggingFaceM4/idefics3
It's a multimodal model based on Llama 3.1 that accepts an arbitrary number of interleaved images with text with a huge context window (10k tokens!) β¨
Supported by Hugging Face transformers π€
- 2 replies
reacted to
dvilasuero's
post with π€ππ₯
7 months ago
Post
8073
Today is a huge day in Argillaβs history. We couldnβt be more excited to share this with the community: weβre joining Hugging Face!
Weβre embracing a larger mission, becoming part of a brilliant and kind team and a shared vision about the future of AI.
Over the past year, weβve been collaborating with Hugging Face on countless projects: launching partner of Docker Spaces, empowering the community to clean Alpaca translations into Spanish and other languages, launching argilla/notus-7b-v1 building on Zephyrβs learnings, the Data is Better Together initiative with hundreds of community contributors, or releasing argilla/OpenHermesPreferences, one of the largest open preference tuning datasets
After more than 2,000 Slack messages and over 60 people collaborating for over a year, it already felt like we were part of the same team, pushing in the same direction. After a week of the smoothest transition you can imagine, weβre now the same team.
To those of you whoβve been following us, this wonβt be a huge surprise, but it will be a big deal in the coming months. This acquisition means weβll double down on empowering the community to build and collaborate on high quality datasets, weβll bring full support for multimodal datasets, and weβll be in a better place to collaborate with the Open Source AI community. For enterprises, this means that the Enterprise Hub will unlock highly requested features like single sign-on and integration with Inference Endpoints.
As a founder, I am proud of the Argilla team. We're now part of something bigger and a larger team but with the same values, culture, and goals. Grateful to have shared this journey with my beloved co-founders Paco and AmΓ©lie.
Finally, huge thanks to the Chief Llama Officer @osanseviero for sparking this and being such a great partner during the acquisition process.
Would love to answer any questions you have so feel free to add them below!
Weβre embracing a larger mission, becoming part of a brilliant and kind team and a shared vision about the future of AI.
Over the past year, weβve been collaborating with Hugging Face on countless projects: launching partner of Docker Spaces, empowering the community to clean Alpaca translations into Spanish and other languages, launching argilla/notus-7b-v1 building on Zephyrβs learnings, the Data is Better Together initiative with hundreds of community contributors, or releasing argilla/OpenHermesPreferences, one of the largest open preference tuning datasets
After more than 2,000 Slack messages and over 60 people collaborating for over a year, it already felt like we were part of the same team, pushing in the same direction. After a week of the smoothest transition you can imagine, weβre now the same team.
To those of you whoβve been following us, this wonβt be a huge surprise, but it will be a big deal in the coming months. This acquisition means weβll double down on empowering the community to build and collaborate on high quality datasets, weβll bring full support for multimodal datasets, and weβll be in a better place to collaborate with the Open Source AI community. For enterprises, this means that the Enterprise Hub will unlock highly requested features like single sign-on and integration with Inference Endpoints.
As a founder, I am proud of the Argilla team. We're now part of something bigger and a larger team but with the same values, culture, and goals. Grateful to have shared this journey with my beloved co-founders Paco and AmΓ©lie.
Finally, huge thanks to the Chief Llama Officer @osanseviero for sparking this and being such a great partner during the acquisition process.
Would love to answer any questions you have so feel free to add them below!
- 28 replies
reacted to
tomaarsen's
post with π₯β€οΈ
7 months ago
Post
1984
βΌοΈSentence Transformers v3.0 is out! You can now train and finetune embedding models with multi-GPU training, bf16 support, loss logging, callbacks & much more. I also release 50+ datasets to train on.
1οΈβ£ Training Refactor
Embedding models can now be trained using an extensive trainer with a lot of powerful features:
- MultiGPU Training (Data Parallelism (DP) and Distributed Data Parallelism (DDP))
- bf16 training support; loss logging
- Evaluation datasets + evaluation loss
- Improved callback support + an excellent Weights & Biases integration
- Gradient checkpointing, gradient accumulation
- Improved model card generation
- Resuming from a training checkpoint without performance loss
- Hyperparameter Optimization
and much more!
Read my detailed blogpost to learn about the components that make up this new training approach: https://huggingface.co./blog/train-sentence-transformers
2οΈβ£ Similarity Score
Not sure how to compare embeddings? Don't worry, you can now use
3οΈβ£ Additional Kwargs
Sentence Transformers relies on various Transformers instances (AutoModel, AutoTokenizer, AutoConfig), but it was hard to provide valuable keyword arguments to these (like 'torch_dtype=torch.bfloat16' to load a model a lower precision for 2x inference speedup). This is now easy!
4οΈβ£ Hyperparameter Optimization
Sentence Transformers now ships with HPO, allowing you to effectively choose your hyperparameters for your data and task.
5οΈβ£ Dataset Release
To help you out with finetuning models, I've released 50+ ready-to-go datasets that can be used with training or finetuning embedding models: sentence-transformers/embedding-model-datasets-6644d7a3673a511914aa7552
Full release notes: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.0.0
1οΈβ£ Training Refactor
Embedding models can now be trained using an extensive trainer with a lot of powerful features:
- MultiGPU Training (Data Parallelism (DP) and Distributed Data Parallelism (DDP))
- bf16 training support; loss logging
- Evaluation datasets + evaluation loss
- Improved callback support + an excellent Weights & Biases integration
- Gradient checkpointing, gradient accumulation
- Improved model card generation
- Resuming from a training checkpoint without performance loss
- Hyperparameter Optimization
and much more!
Read my detailed blogpost to learn about the components that make up this new training approach: https://huggingface.co./blog/train-sentence-transformers
2οΈβ£ Similarity Score
Not sure how to compare embeddings? Don't worry, you can now use
model.similarity(embeddings1, embeddings2) and you'll get your similarity scores immediately. Model authors can specify their desired similarity score, so you don't have to worry about it anymore!3οΈβ£ Additional Kwargs
Sentence Transformers relies on various Transformers instances (AutoModel, AutoTokenizer, AutoConfig), but it was hard to provide valuable keyword arguments to these (like 'torch_dtype=torch.bfloat16' to load a model a lower precision for 2x inference speedup). This is now easy!
4οΈβ£ Hyperparameter Optimization
Sentence Transformers now ships with HPO, allowing you to effectively choose your hyperparameters for your data and task.
5οΈβ£ Dataset Release
To help you out with finetuning models, I've released 50+ ready-to-go datasets that can be used with training or finetuning embedding models: sentence-transformers/embedding-model-datasets-6644d7a3673a511914aa7552
Full release notes: https://github.com/UKPLab/sentence-transformers/releases/tag/v3.0.0
reacted to
merve's
post with π
7 months ago
Post
1758
New open Vision Language Model by
@Google
: PaliGemma ππ€
π Comes in 3B, pretrained, mix and fine-tuned models in 224, 448 and 896 resolution
𧩠Combination of Gemma 2B LLM and SigLIP image encoder
π€ Supported in transformers
PaliGemma can do..
𧩠Image segmentation and detection! π€―
π Detailed document understanding and reasoning
π Visual question answering, captioning and any other VLM task!
Read our blog π hf.co/blog/paligemma
Try the demo πͺ hf.co/spaces/google/paligemma
Check out the Spaces and the models all in the collection π google/paligemma-release-6643a9ffbf57de2ae0448dda
Collection of fine-tuned PaliGemma models google/paligemma-ft-models-6643b03efb769dad650d2dda
π Comes in 3B, pretrained, mix and fine-tuned models in 224, 448 and 896 resolution
𧩠Combination of Gemma 2B LLM and SigLIP image encoder
π€ Supported in transformers
PaliGemma can do..
𧩠Image segmentation and detection! π€―
π Detailed document understanding and reasoning
π Visual question answering, captioning and any other VLM task!
Read our blog π hf.co/blog/paligemma
Try the demo πͺ hf.co/spaces/google/paligemma
Check out the Spaces and the models all in the collection π google/paligemma-release-6643a9ffbf57de2ae0448dda
Collection of fine-tuned PaliGemma models google/paligemma-ft-models-6643b03efb769dad650d2dda
- 13 replies
posted
an
update
8 months ago
Post
3098
π₯ Prometheus 2 was recently released by Kaist AI as an alternative and closely mirroring both human and GPT-4 evaluation, and surpassing the former Prometheus!
prometheus-eval/prometheus-7b-v2.0
prometheus-eval/prometheus-8x7b-v2.0
π¬οΈFine-tuned on top of mistralai/Mistral-7B-Instruct-v0.2 and mistralai/Mixtral-8x7B-Instruct-v0.1
ποΈThe datasets used for fine-tuning have been publicly released i.e. prometheus-eval/Feedback-Collection and prometheus-eval/Preference-Collection
π€π»Unified LM evaluator for absolute (a single prompt-completion pair) and relative (two completions for a given prompt) due to model merging
βNo longer needs a mandatory reference / golden answer, but can still be provided optionally
πSurpasses the former version of Prometheus, and has a high correlation with human, GPT-4, and Claude 3 Opus scores when evaluating LMs
πApache 2.0 license
Long-story short, an amazing job from Kaist AI bridging the gap with LLM evaluators other than proprietary and bigger models!
This week at Argilla, we decided to add a new task to use Prometheus 2 as an LLM evaluator using
π± Using
Find the generated dataset and the code at distilabel-internal-testing/instruction-dataset-prometheus
prometheus-eval/prometheus-7b-v2.0
prometheus-eval/prometheus-8x7b-v2.0
π¬οΈFine-tuned on top of mistralai/Mistral-7B-Instruct-v0.2 and mistralai/Mixtral-8x7B-Instruct-v0.1
ποΈThe datasets used for fine-tuning have been publicly released i.e. prometheus-eval/Feedback-Collection and prometheus-eval/Preference-Collection
π€π»Unified LM evaluator for absolute (a single prompt-completion pair) and relative (two completions for a given prompt) due to model merging
βNo longer needs a mandatory reference / golden answer, but can still be provided optionally
πSurpasses the former version of Prometheus, and has a high correlation with human, GPT-4, and Claude 3 Opus scores when evaluating LMs
πApache 2.0 license
Long-story short, an amazing job from Kaist AI bridging the gap with LLM evaluators other than proprietary and bigger models!
This week at Argilla, we decided to add a new task to use Prometheus 2 as an LLM evaluator using
distilabel, so we implemented PrometheusEval.π± Using
PrometheusEval running their 7B variant with vLLM in a single L40 on top of
HuggingFaceH4/instruction-dataset, we got the 327 existing prompt-completion pairs evaluated and pushed to the Hub in less than 2 minutes!Find the generated dataset and the code at distilabel-internal-testing/instruction-dataset-prometheus
- 1 reply
reacted to
davanstrien's
post with π€
8 months ago
Post
1671
As part of the Data is Better Together MPEP project, we are now at the point where some translation efforts have successfully translated 500 highly ranked prompts into a new target language (amazing work from
@Rijgersberg
et al!)
Our next step is to use these translated prompts to evaluate the performance of LLMs for non English languages.
Does LLM, as a judge, work outside of English?
Ideally, it would be compelling to leverage LLMs to judge models for non-English since this significantly lowers the barrier to evaluating models (although it doesn't remove this barrier altogether).
What we want to know is:
- does auto/LLM eval work in general for a particular language
- which model(s) works best as a judge
- do LLMs' judgments of non-English models match human preferences?
We're starting to think about how to approach this. If you have any ideas of possible approaches feel free to comment or join the discussion here: https://github.com/huggingface/data-is-better-together/issues/61
Other ideas...
Could an approach like Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models (2404.18796) with the SOA models for a particular language work? i.e., choose 4 of the best open LLMs for Arabic and use those at the pool of raters rather than relying on one powerful judge LLM?
Our next step is to use these translated prompts to evaluate the performance of LLMs for non English languages.
Does LLM, as a judge, work outside of English?
Ideally, it would be compelling to leverage LLMs to judge models for non-English since this significantly lowers the barrier to evaluating models (although it doesn't remove this barrier altogether).
What we want to know is:
- does auto/LLM eval work in general for a particular language
- which model(s) works best as a judge
- do LLMs' judgments of non-English models match human preferences?
We're starting to think about how to approach this. If you have any ideas of possible approaches feel free to comment or join the discussion here: https://github.com/huggingface/data-is-better-together/issues/61
Other ideas...
Could an approach like Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models (2404.18796) with the SOA models for a particular language work? i.e., choose 4 of the best open LLMs for Arabic and use those at the pool of raters rather than relying on one powerful judge LLM?
reacted to
davanstrien's
post with π€
8 months ago
Post
1538
Can you create domain-specific synthetic datasets in under 20 minutes?
@burtenshaw recently launched the Domain Specific Dataset Project as part of Data is Better Together. As part of this, Ben created a Space that you can use to define some key perspectives and concepts from a domain. This seed dataset can then be used to generate a synthetic dataset for a particular domain.
In less than 30 minutes this afternoon, I created a domain-specific dataset focused on data-centric machine learning using these tools: davanstrien/data-centric-ml-sft.
You can create your own domain specific datasets using this approach. Find the steps to follow here: https://github.com/huggingface/data-is-better-together/blob/main/domain-specific-datasets/README.md
@burtenshaw recently launched the Domain Specific Dataset Project as part of Data is Better Together. As part of this, Ben created a Space that you can use to define some key perspectives and concepts from a domain. This seed dataset can then be used to generate a synthetic dataset for a particular domain.
In less than 30 minutes this afternoon, I created a domain-specific dataset focused on data-centric machine learning using these tools: davanstrien/data-centric-ml-sft.
You can create your own domain specific datasets using this approach. Find the steps to follow here: https://github.com/huggingface/data-is-better-together/blob/main/domain-specific-datasets/README.md
-
- 1 reply
reacted to
tomaarsen's
post with π₯
8 months ago
Post
3172
π Sentence Transformers v2.7.0 is out! Featuring a new loss function, easier Matryoshka model inference & evaluation, CrossEncoder improvements & Intel Gaudi2 Accelerator support. Details:
1οΈβ£ A new loss function: CachedGISTEmbedLoss
This loss function is a combination of CachedMultipleNegativesRankingLoss and the GISTEmbedLoss, both of which are already excellent. The caching mechanism allows for much higher batch sizes with constant memory usage, which boosts training performance. The GIST part introduces a guide model to guide the in-batch negative sample selection. This prevents false negatives, resulting in a stronger training signal.
2οΈβ£ Automatic Matryoshka model truncation
Matryoshka models produce embeddings that are still useful after truncation. However, this truncation always had to be done manually, until now! We've added a
3οΈβ£ Additionally, you can now specify
4οΈβ£ CrossEncoder improvements
The CrossEncoder now supports 'push_to_hub' to upload trained reranker models to Hugging Face. Additionally, CrossEncoders now support
5οΈβ£ Inference on Intel Gaudi2
If you have an Intel Gaudi2 Accelerator, Sentence Transformers now uses it automatically for even faster inference. No changes are necessary to your code, the device is automatically detected!
Check out the release notes for all of the details: https://github.com/UKPLab/sentence-transformers/releases/tag/v2.7.0
I'm very excited for the upcoming releases: I'm making great progress with a notable v3 refactor that should heavily improve the training process for embedding models!
1οΈβ£ A new loss function: CachedGISTEmbedLoss
This loss function is a combination of CachedMultipleNegativesRankingLoss and the GISTEmbedLoss, both of which are already excellent. The caching mechanism allows for much higher batch sizes with constant memory usage, which boosts training performance. The GIST part introduces a guide model to guide the in-batch negative sample selection. This prevents false negatives, resulting in a stronger training signal.
2οΈβ£ Automatic Matryoshka model truncation
Matryoshka models produce embeddings that are still useful after truncation. However, this truncation always had to be done manually, until now! We've added a
truncate_dim option to the Sentence Transformer constructor. This also allows truncation when using HuggingFaceEmbeddings from LlamaIndex or LangChain.3οΈβ£ Additionally, you can now specify
truncate_dim in evaluators to get the performance after truncation. (Hint: it's surprisingly good, even for models not trained with MatryoshkaLoss, and it can speed up e.g. clustering, retrieval, etc.)4οΈβ£ CrossEncoder improvements
The CrossEncoder now supports 'push_to_hub' to upload trained reranker models to Hugging Face. Additionally, CrossEncoders now support
trust_remote_code to load models with custom modelling code.5οΈβ£ Inference on Intel Gaudi2
If you have an Intel Gaudi2 Accelerator, Sentence Transformers now uses it automatically for even faster inference. No changes are necessary to your code, the device is automatically detected!
Check out the release notes for all of the details: https://github.com/UKPLab/sentence-transformers/releases/tag/v2.7.0
I'm very excited for the upcoming releases: I'm making great progress with a notable v3 refactor that should heavily improve the training process for embedding models!
- 2 replies
reacted to
sayakpaul's
post with π€π₯
8 months ago
Post
2901
We're introducing experimental support for
If you have multiple GPUs you want to use to distribute the pipeline models, you can do so. Additionally, this becomes more useful when you have multiple low-VRAM GPUs.
Documentation:
https://huggingface.co./docs/diffusers/main/en/training/distributed_inference#device-placement
π¨ Currently, only "balanced" device mapping strategy is supported.
device_map in Diffusers π€If you have multiple GPUs you want to use to distribute the pipeline models, you can do so. Additionally, this becomes more useful when you have multiple low-VRAM GPUs.
Documentation:
https://huggingface.co./docs/diffusers/main/en/training/distributed_inference#device-placement
π¨ Currently, only "balanced" device mapping strategy is supported.