SmolvencoderOrg (Smolvencoder)

andito

posted an update 17 days ago

Post

2725

Many VLMs claim to process hours of video. But can they follow the story?🤔
Today, we introduce TimeScope: The benchmark that separates true temporal understanding from marketing hype. Let's see how much VLMs really understand!⏳

We test three skills that matter for real-world use:
🔎 Localized Retrieval: Find a specific action.
🧩 Information Synthesis: Piece together scattered clues.
🏃 Fine-Grained Perception: Analyze detailed motion (e.g., count how many times a person swings an axe).

The results are in, and they're revealing. Only Gemini 2.5 pro handles 1-hour-long videos.
Performance drops sharply with duration, proving that long video understanding is still challenging. We've found the breaking points—now the community can start fixing them.📈

Want to learn more? TimeScope is 100% open-source. Benchmark your model and help us build the next generation of video AI.

📖 Blog:
https://huggingface.co/blog/timescope-video-lmm-benchmark
👩‍💻 Leaderboard & Demo: Apollo-LMMs/TimeScope
📊 Dataset: Apollo-LMMs/TimeScope
⚙️ Eval Code: https://github.com/EvolvingLMMs-Lab/lmms-eval

manu

authored a paper about 1 month ago

Should We Still Pretrain Encoders with Masked Language Modeling?

Paper • 2507.00994 • Published Jul 1 • 75

andito

posted an update about 1 month ago

Post

3979

🧠👁️ Can AI visualize solutions?

Humans often solve visual problems by sketching ideas in our minds. What if Vision-Language Models (VLMs) could do something similar, not by generating full images, but by using internal “mental sketches”?

That’s the idea behind Mirage, a new framework that empowers VLMs to reason using latent visual tokens. Instead of just thinking in words, Mirage mixes in abstract visual representations that help the model solve complex tasks.

These aren't photorealistic images. They're compact, internal representations optimized purely to support reasoning.

🔧 Mirage is trained in two phases:

1) Grounding: It learns to produce latent tokens anchored in real images.
2) Refinement: The model drops the images and learns to generate visual tokens on its own.

📈 And yes, it works!
On challenging benchmarks like Visual Spatial Planning, Jigsaw puzzles, and Spatial Attention Tasks, Mirage clearly outperforms GPT-4o and other strong baselines.
Smart sketches > empty words.

By mimicking the way humans visualize solutions, Mirage gives AI a new kind of imagination, one that’s faster, more efficient, and more human-like.
Kudos to the teams at UMass Amherst and MIT behind this exciting work.
Check the paper: Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens (2506.17218)

4 replies

·

andito

authored a paper 2 months ago

SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics

Paper • 2506.01844 • Published Jun 2 • 126

manu

authored a paper 2 months ago

ViDoRe Benchmark V2: Raising the Bar for Visual Retrieval

Paper • 2505.17166 • Published May 22

andito

authored a paper 4 months ago

SmolVLM: Redefining small and efficient multimodal models

Paper • 2504.05299 • Published Apr 7 • 197

andito

authored a paper 5 months ago

SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion

Paper • 2503.11576 • Published Mar 14 • 114

manu

authored 2 papers 5 months ago

EuroBERT: Scaling Multilingual Encoders for European Languages

Paper • 2503.05500 • Published Mar 7 • 81

MMTEB: Massive Multilingual Text Embedding Benchmark

Paper • 2502.13595 • Published Feb 19 • 38

andito

posted an update 5 months ago

Post

2982

Extremely bullish on @CohereForAI 's Aya Vision (8B & 32B) - new SOTA open-weight VLMs

- 8B wins up to 81% of the time in its class, better than Gemini Flash
- 32B beats Llama 3.2 90B!
- Covers 23 languages, excels in image captioning, VQA & more
- Integrated on transformers from Day 0!

Efficient multimodal models are here to stay!!🔥
Check out their blog! https://huggingface.co/blog/aya-vision

andito

authored a paper 6 months ago

SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model

Paper • 2502.02737 • Published Feb 4 • 241

andito

posted an update 7 months ago

Post

1663

𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝘁𝗵𝗲 𝘄𝗼𝗿𝗹𝗱'𝘀 𝘀𝗺𝗮𝗹𝗹𝗲𝘀𝘁 𝘃𝗶𝘀𝗶𝗼𝗻 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗺𝗼𝗱𝗲𝗹!

We’re thrilled to share 𝗦𝗺𝗼𝗹𝗩𝗟𝗠 (256M & 500M)—the smallest Visual Language Models ever built. Think: running on <1GB of GPU memory—you can fine-tune it on your laptop and run it on your toaster!

Why It’s Game-Changing:
- 𝗢𝘂𝘁𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝘀 𝗟𝗮𝗿𝗴𝗲𝗿 𝗠𝗼𝗱𝗲𝗹𝘀: Even the 256M model surpasses our SOTA 80B-parameter model from just 17 months ago. Over 300x reduction!
𝗠𝗶𝗴𝗵𝘁𝘆 𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆: The 256M version delivers 80% of our 2.2B model’s performance, and the 500M version hits 90%
𝗟𝗶𝗴𝗵𝘁𝗻𝗶𝗻𝗴-𝗙𝗮𝘀𝘁 𝗦𝗲𝗮𝗿𝗰𝗵: SmolVLM integrates with ColiPali for state-of-the-art retrieval speeds—on par with models 10x bigger. That means cheaper, faster indexing and real-world impact.

What’s New Under the Hood:
- 𝗡𝗲𝘄 𝗩𝗶𝘀𝗶𝗼𝗻 𝗘𝗻𝗰𝗼𝗱𝗲𝗿: Smaller overall size (400M -> 93M), but with higher resolution.
- 𝗛𝗶𝗴𝗵𝗲𝗿 𝗣𝗶𝘅𝗲𝗹𝘀/𝗧𝗼𝗸𝗲𝗻: 4096 vs. 1820—more efficient image processing.
- 𝗦𝗺𝗮𝗿𝘁 𝗧𝗼𝗸𝗲𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻: Faster training and a performance boost.

Check our blog: https://huggingface.co/blog/smolervlm
The models: HuggingFaceTB/smolvlm-256m-and-500m-6791fafc5bb0ab8acc960fb0
The demo: HuggingFaceTB/SmolVLM-256M-Demo

1 reply

·

andito

posted an update 8 months ago

Post

1987

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 !

andito

posted an update 9 months ago

Post

3409

Let's go! We are releasing SmolVLM, a smol 2B VLM built for on-device inference that outperforms all models at similar GPU RAM usage and tokens throughputs.

- SmolVLM generates tokens 7.5 to 16 times faster than Qwen2-VL! 🤯
- Other models at this size crash a laptop, but SmolVLM comfortably generates 17 tokens/sec on a macbook! 🚀
- SmolVLM can be fine-tuned on a Google collab! Or process millions of documents with a consumer GPU!
- SmolVLM even outperforms larger models in video benchmarks, despite not even being trained on videos!

Check out more!
Demo: HuggingFaceTB/SmolVLM
Blog: https://huggingface.co/blog/smolvlm
Model: HuggingFaceTB/SmolVLM-Instruct
Fine-tuning script: https://github.com/huggingface/smollm/blob/main/finetuning/Smol_VLM_FT.ipynb

manu

authored a paper 10 months ago

EuroLLM: Multilingual Language Models for Europe

Paper • 2409.16235 • Published Sep 24, 2024 • 28

andito

posted an update 11 months ago

Post

1100

Hugging face presents FineVideo 🎥! Unlocking the next generation of Video understanding 🚀

🤯3400 hours of annotated Creative Common videos with rich character descriptions, scene splits, mood, and content descriptions per scene as well as QA pairs.
🔥
@mfarre processed over 2M videos of Youtube-CC to make this incredibly powerful selection.

Very psyched to fine-tune idefics on this dataset. ⚡️
Explore the videos: HuggingFaceFV/FineVideo-Explorer

andito

authored 2 papers 11 months ago

GACELA -- A generative adversarial context encoder for long audio inpainting

Paper • 2005.05032 • Published May 11, 2020

Adversarial Generation of Time-Frequency Features with application in audio synthesis

Paper • 1902.04072 • Published Feb 11, 2019

andito

posted an update 11 months ago

Post

1653

🚀 Introducing Hugging Face's Multilingual Speech-to-Speech! 🎤
💬Our modular, cross-platform pipeline to run GPT4o-like experiences on device can now seamlessly switch languages mid-conversation with an imperceptible 100ms delay.

🌟 Building on an amazing early reception with 2600 stars on GitHub 🌟
🚀 We are expanding the library to support multiple languages
🔥 Try it out with a flag: --language fr
🤯 Or don't set the flag and let the system detect the language

💡 What feature should we add next?

1 reply

·

andito

authored a paper 12 months ago

Building and better understanding vision-language models: insights and future directions

Paper • 2408.12637 • Published Aug 22, 2024 • 132

Smolvencoder

AI & ML interests

Recent Activity

Should We Still Pretrain Encoders with Masked Language Modeling?

SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics

ViDoRe Benchmark V2: Raising the Bar for Visual Retrieval

SmolVLM: Redefining small and efficient multimodal models

SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion

EuroBERT: Scaling Multilingual Encoders for European Languages

MMTEB: Massive Multilingual Text Embedding Benchmark

SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model

EuroLLM: Multilingual Language Models for Europe

GACELA -- A generative adversarial context encoder for long audio inpainting

Adversarial Generation of Time-Frequency Features with application in audio synthesis

Building and better understanding vision-language models: insights and future directions

AI & ML interests

Recent Activity

Team members 3

SmolvencoderOrg's activity