HuggingFaceTB/SmolLM2-SmolTalk-ep1
Text Generation
•
Updated
Hugging Face TB Research
Enterprise
community
AI & ML interests
Exploring smol models and high quality web and synthetic datasets, generated by LLMs (TB is for Textbook, as inspired by the "Textbooks are all your need" paper)
Recent Activity
View all activity
HuggingFaceTB's activity
loubnabnl
updated
a
model
about 5 hours ago
davanstrien
posted
an
update
about 22 hours ago
Post
722
Why choose between strong LLM reasoning and efficient models?
Use DeepSeek to generate high-quality training data, then distil that knowledge into ModernBERT answerdotai/ModernBERT-base for fast, efficient classification.
Blog post: https://danielvanstrien.xyz/posts/2025/deepseek/distil-deepseek-modernbert.html
Use DeepSeek to generate high-quality training data, then distil that knowledge into ModernBERT answerdotai/ModernBERT-base for fast, efficient classification.
Blog post: https://danielvanstrien.xyz/posts/2025/deepseek/distil-deepseek-modernbert.html
cfahlgren1
posted
an
update
1 day ago
Post
1186
If you haven't seen yet, we just released Inference Providers 🔀
> 4 new serverless inference providers on the Hub 🤯
> Use your HF API key or personal key with all providers 🔑
> Chat with Deepseek R1, V3, and more on HF Hub 🐋
> We support Sambanova, TogetherAI, Replicate, and Fal.ai 💪
Best of all, we don't charge any markup on top of the provider 🫰 Have you tried it out yet? HF Pro accounts get $2 of free usage for the provider inference.
> 4 new serverless inference providers on the Hub 🤯
> Use your HF API key or personal key with all providers 🔑
> Chat with Deepseek R1, V3, and more on HF Hub 🐋
> We support Sambanova, TogetherAI, Replicate, and Fal.ai 💪
Best of all, we don't charge any markup on top of the provider 🫰 Have you tried it out yet? HF Pro accounts get $2 of free usage for the provider inference.
Post
1213
🚀 The open source community is unstoppable: 4M total downloads for DeepSeek models on Hugging Face, with 3.2M coming from the +600 models created by the community.
That's 30% more than yesterday!
That's 30% more than yesterday!
- 1 reply
davanstrien
posted
an
update
2 days ago
Post
1526
Updated the ColPali Query Generator Space
davanstrien/ColPali-Query-Generator to use
Qwen/Qwen2.5-VL-7B-Instruct.
Given an input image, it generates several queries along with explanations to justify them. This approach can generate synthetic data for fine-tuning ColPali models.
Given an input image, it generates several queries along with explanations to justify them. This approach can generate synthetic data for fine-tuning ColPali models.
Post
7158
Yes, DeepSeek R1's release is impressive. But the real story is what happened in just 7 days after:
- Original release: 8 models, 540K downloads. Just the beginning...
- The community turned those open-weight models into +550 NEW models on Hugging Face. Total downloads? 2.5M—nearly 5X the originals.
The reason? DeepSeek models are open-weight, letting anyone build on top of them. Interesting to note that the community focused on quantized versions for better efficiency & accessibility. They want models that use less memory, run faster, and are more energy-efficient.
When you empower builders, innovation explodes. For everyone. 🚀
The most popular community model? @bartowski 's DeepSeek-R1-Distill-Qwen-32B-GGUF version — 1M downloads alone.
- Original release: 8 models, 540K downloads. Just the beginning...
- The community turned those open-weight models into +550 NEW models on Hugging Face. Total downloads? 2.5M—nearly 5X the originals.
The reason? DeepSeek models are open-weight, letting anyone build on top of them. Interesting to note that the community focused on quantized versions for better efficiency & accessibility. They want models that use less memory, run faster, and are more energy-efficient.
When you empower builders, innovation explodes. For everyone. 🚀
The most popular community model? @bartowski 's DeepSeek-R1-Distill-Qwen-32B-GGUF version — 1M downloads alone.
- 4 replies
davanstrien
posted
an
update
3 days ago
Post
1865
🌍 Big step for multilingual AI data!
The Hugging Face community has rated educational content in languages spoken by 1.6 billion people! New additions:
• Japanese
• Italian
• Old High German
Learn more and contribute: https://huggingface.co/blog/davanstrien/fineweb2-community
These ratings can help enhance training data for major world languages.
The Hugging Face community has rated educational content in languages spoken by 1.6 billion people! New additions:
• Japanese
• Italian
• Old High German
Learn more and contribute: https://huggingface.co/blog/davanstrien/fineweb2-community
These ratings can help enhance training data for major world languages.
-
- 1 reply
davidberenstein1957
posted
an
update
3 days ago
Post
1684
Let's uncover the post-training dataset from DeepSeek-R1 with Magpie!
Pass pre-query tokens
We can get realistic examples!
Gist: https://gist.github.com/davidberenstein1957/3f20046ce57395a6aba13f8b4e956b59
Pass pre-query tokens
<｜begin▁of▁sentence｜>User: , let the model generate the rest.We can get realistic examples!
Gist: https://gist.github.com/davidberenstein1957/3f20046ce57395a6aba13f8b4e956b59
-
- 6 replies
Post
9358
We are reproducing the full DeepSeek R1 data and training pipeline so everybody can use their recipe. Instead of doing it in secret we can do it together in the open!
🧪 Step 1: replicate the R1-Distill models by distilling a high-quality reasoning corpus from DeepSeek-R1.
🧠 Step 2: replicate the pure RL pipeline that DeepSeek used to create R1-Zero. This will involve curating new, large-scale datasets for math, reasoning, and code.
🔥 Step 3: show we can go from base model -> SFT -> RL via multi-stage training.
Follow along: https://github.com/huggingface/open-r1
🧪 Step 1: replicate the R1-Distill models by distilling a high-quality reasoning corpus from DeepSeek-R1.
🧠 Step 2: replicate the pure RL pipeline that DeepSeek used to create R1-Zero. This will involve curating new, large-scale datasets for math, reasoning, and code.
🔥 Step 3: show we can go from base model -> SFT -> RL via multi-stage training.
Follow along: https://github.com/huggingface/open-r1
- 5 replies
Post
4317
Oof, what a week! 🥵 So many things have happened, let's recap!
merve/jan-24-releases-6793d610774073328eac67a9
Multimodal 💬
- We have released SmolVLM -- tiniest VLMs that come in 256M and 500M, with it's retrieval models ColSmol for multimodal RAG 💗
- UI-TARS are new models by ByteDance to unlock agentic GUI control 🤯 in 2B, 7B and 72B
- Alibaba DAMO lab released VideoLlama3, new video LMs that come in 2B and 7B
- MiniMaxAI released Minimax-VL-01, where decoder is based on MiniMax-Text-01 456B MoE model with long context
- Dataset: Yale released a new benchmark called MMVU
- Dataset: CAIS released Humanity's Last Exam (HLE) a new challenging MM benchmark
LLMs 📖
- DeepSeek-R1 & DeepSeek-R1-Zero: gigantic 660B reasoning models by DeepSeek, and six distilled dense models, on par with o1 with MIT license! 🤯
- Qwen2.5-Math-PRM: new math models by Qwen in 7B and 72B
- NVIDIA released AceMath and AceInstruct, new family of models and their datasets (SFT and reward ones too!)
Audio 🗣️
- Llasa is a new speech synthesis model based on Llama that comes in 1B,3B, and 8B
- TangoFlux is a new audio generation model trained from scratch and aligned with CRPO
Image/Video/3D Generation ⏯️
- Flex.1-alpha is a new 8B pre-trained diffusion model by ostris similar to Flux
- tencent released Hunyuan3D-2, new 3D asset generation from images
Multimodal 💬
- We have released SmolVLM -- tiniest VLMs that come in 256M and 500M, with it's retrieval models ColSmol for multimodal RAG 💗
- UI-TARS are new models by ByteDance to unlock agentic GUI control 🤯 in 2B, 7B and 72B
- Alibaba DAMO lab released VideoLlama3, new video LMs that come in 2B and 7B
- MiniMaxAI released Minimax-VL-01, where decoder is based on MiniMax-Text-01 456B MoE model with long context
- Dataset: Yale released a new benchmark called MMVU
- Dataset: CAIS released Humanity's Last Exam (HLE) a new challenging MM benchmark
LLMs 📖
- DeepSeek-R1 & DeepSeek-R1-Zero: gigantic 660B reasoning models by DeepSeek, and six distilled dense models, on par with o1 with MIT license! 🤯
- Qwen2.5-Math-PRM: new math models by Qwen in 7B and 72B
- NVIDIA released AceMath and AceInstruct, new family of models and their datasets (SFT and reward ones too!)
Audio 🗣️
- Llasa is a new speech synthesis model based on Llama that comes in 1B,3B, and 8B
- TangoFlux is a new audio generation model trained from scratch and aligned with CRPO
Image/Video/3D Generation ⏯️
- Flex.1-alpha is a new 8B pre-trained diffusion model by ostris similar to Flux
- tencent released Hunyuan3D-2, new 3D asset generation from images
-
- 7 replies
Post
2083
smolagents can see 🔥
we just shipped vision support to smolagents 🤗 agentic computers FTW
you can now:
💻 let the agent get images dynamically (e.g. agentic web browser)
📑 pass images at the init of the agent (e.g. chatting with documents, filling forms automatically etc)
with few LoC change! 🤯
you can use transformers models locally (like Qwen2VL) OR plug-in your favorite multimodal inference provider (gpt-4o, antrophic & co) 🤠
read our blog http://hf.co/blog/smolagents-can-see
we just shipped vision support to smolagents 🤗 agentic computers FTW
you can now:
💻 let the agent get images dynamically (e.g. agentic web browser)
📑 pass images at the init of the agent (e.g. chatting with documents, filling forms automatically etc)
with few LoC change! 🤯
you can use transformers models locally (like Qwen2VL) OR plug-in your favorite multimodal inference provider (gpt-4o, antrophic & co) 🤠
read our blog http://hf.co/blog/smolagents-can-see
Post
1443
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
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
Post
1720
What's at stake with Meta's decision to change its content moderation policy?
@giadap
has, by far, the most thoughtful take I’ve seen on this question. Read her op-ed: https://www.techpolicy.press/when-freedom-bites-back-meta-moderation-and-the-limits-of-tolerance/
davidberenstein1957
posted
an
update
9 days ago
Post
1849
The RAG's in the bag!
You can now use the Synthetic Data Generator with your own domain-specific seed data to generate a dataset for fine-tuning retrieval or reranking models.
GitHub: https://buff.ly/49IDSmd
Spaces: https://buff.ly/3Y1S99z
Blog: https://huggingface.co/blog/sdiazlor/fine-tune-modernbert-for-rag-with-synthetic-data
You can now use the Synthetic Data Generator with your own domain-specific seed data to generate a dataset for fine-tuning retrieval or reranking models.
GitHub: https://buff.ly/49IDSmd
Spaces: https://buff.ly/3Y1S99z
Blog: https://huggingface.co/blog/sdiazlor/fine-tune-modernbert-for-rag-with-synthetic-data
- 1 reply
Post
573
Fun fact: you can get any DeepSeek-R1-Qwen **abliterated** by using one of these LoRA adapters (GGUF available!)
ngxson/extracted-lora-mergekit-677d5c3eea0b6a7661201846
ngxson/extracted-lora-mergekit-677d5c3eea0b6a7661201846
Post
1799
Reminder: Don’t. Use. ChatGPT. As. A. Calculator. Seriously. 🤖
Loved listening to @sasha on Hard Fork—it really made me think.
A few takeaways that hit home:
- Individual culpability only gets you so far. The real priority: demanding accountability and transparency from companies.
- Evaluate if generative AI is the right tool for certain tasks (like search) before using it.
Curious about the full conversation? https://www.nytimes.com/2025/01/17/podcasts/hardfork-tiktok-rednote-environment.html. Give it a listen—it’s worth it! 🌍
Loved listening to @sasha on Hard Fork—it really made me think.
A few takeaways that hit home:
- Individual culpability only gets you so far. The real priority: demanding accountability and transparency from companies.
- Evaluate if generative AI is the right tool for certain tasks (like search) before using it.
Curious about the full conversation? https://www.nytimes.com/2025/01/17/podcasts/hardfork-tiktok-rednote-environment.html. Give it a listen—it’s worth it! 🌍
- 1 reply
Post
2531
Everything that happened this week in open AI, a recap 🤠
merve/jan-17-releases-678a673a9de4a4675f215bf5
👀 Multimodal
- MiniCPM-o 2.6 is a new sota any-to-any model by OpenBMB
(vision, speech and text!)
- VideoChat-Flash-Qwen2.5-2B is new video multimodal models by OpenGVLab that come in sizes 2B & 7B in resolutions 224 & 448
- ByteDance released larger SA2VA that comes in 26B parameters
- Dataset: VRC-Bench is a new diverse benchmark for multimodal LLM reasoning performance
💬 LLMs
- MiniMax-Text-01 is a new huge language model (456B passive 45.9B active params) by MiniMaxAI with context length of 4M tokens 🤯
- Dataset: Sky-T1-data-17k is a diverse dataset used to train Sky-T1-32B
- kyutai released Helium-1-Preview-2B is a new small multilingual LM
- Wayfarer-12B is a new LLM able to write D&D 🧙🏻♂️
- ReaderLM-v2 is a new HTML parsing model by Jina AI
- Dria released, Dria-Agent-a-3B, new agentic coding model (Pythonic function calling) based on Qwen2.5 Coder
- Unsloth released Phi-4, faster and memory efficient Llama 3.3
🖼️ Vision
- MatchAnything is a new foundation model for matching
- FitDit is a high-fidelity VTON model based on DiT architecture
🗣️ Audio
- OuteTTS-0.3-1B is a new multilingual text-to-speech model with voice cloning and emotion control capabilities
📖 Retrieval
- lightblue released a new reranker based on Qwen2.5 LB-reranker-0.5B-v1.0 that can handle 95+ languages
- cde-small-v2 is a new sota small retrieval model by
@jxm
👀 Multimodal
- MiniCPM-o 2.6 is a new sota any-to-any model by OpenBMB
(vision, speech and text!)
- VideoChat-Flash-Qwen2.5-2B is new video multimodal models by OpenGVLab that come in sizes 2B & 7B in resolutions 224 & 448
- ByteDance released larger SA2VA that comes in 26B parameters
- Dataset: VRC-Bench is a new diverse benchmark for multimodal LLM reasoning performance
💬 LLMs
- MiniMax-Text-01 is a new huge language model (456B passive 45.9B active params) by MiniMaxAI with context length of 4M tokens 🤯
- Dataset: Sky-T1-data-17k is a diverse dataset used to train Sky-T1-32B
- kyutai released Helium-1-Preview-2B is a new small multilingual LM
- Wayfarer-12B is a new LLM able to write D&D 🧙🏻♂️
- ReaderLM-v2 is a new HTML parsing model by Jina AI
- Dria released, Dria-Agent-a-3B, new agentic coding model (Pythonic function calling) based on Qwen2.5 Coder
- Unsloth released Phi-4, faster and memory efficient Llama 3.3
🖼️ Vision
- MatchAnything is a new foundation model for matching
- FitDit is a high-fidelity VTON model based on DiT architecture
🗣️ Audio
- OuteTTS-0.3-1B is a new multilingual text-to-speech model with voice cloning and emotion control capabilities
📖 Retrieval
- lightblue released a new reranker based on Qwen2.5 LB-reranker-0.5B-v1.0 that can handle 95+ languages
- cde-small-v2 is a new sota small retrieval model by
@jxm
davidberenstein1957
posted
an
update
13 days ago
Post
1231
You can now use the "Synthetic Data Generator" at a much larger scale with your preferred inference engine: Ollama, vLLM, TGI, and serverless inference! 🔥
Install, configure, launch!
Space: argilla/synthetic-data-generator
Examples: https://github.com/argilla-io/synthetic-data-generator/tree/main/examples
Install, configure, launch!
Space: argilla/synthetic-data-generator
Examples: https://github.com/argilla-io/synthetic-data-generator/tree/main/examples
Post
1976
New smolagents example landed on Hugging Face cookbook 🤠
Learn how to create an inventory managing multi-agent system with smolagents, MongoDB and DeepSeek Chat 📖 https://huggingface.co/learn/cookbook/mongodb_smolagents_multi_micro_agents
Learn how to create an inventory managing multi-agent system with smolagents, MongoDB and DeepSeek Chat 📖 https://huggingface.co/learn/cookbook/mongodb_smolagents_multi_micro_agents