SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
Paper
β’
2502.02737
β’
Published
β’
62
Florent Daudens
fdaudens
AI & ML interests
AI & Journalism
Recent Activity
upvoted
a
paper
about 2 hours ago
SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language
Model
upvoted
a
paper
about 2 hours ago
Fully Autonomous AI Agents Should Not be Developed
liked
a Space
about 14 hours ago
deepseek-ai/Janus-Pro-7B
Organizations
fdaudens's activity
upvoted
2
papers
about 2 hours ago
liked
a
Space
about 14 hours ago
upvoted
an
article
1 day ago
Article
DABStep: Data Agent Benchmark for Multi-step Reasoning
β’
26
liked
a
Space
1 day ago
upvoted
2
articles
2 days ago
Article
π#86: Four Freedoms of truly open AI
By
and 1 other
β’
β’
5Article
From Hippocrates to AI: Reflections on the Evolution of Consent
By
β’
β’
8
liked
a
Space
2 days ago
upvoted
2
articles
2 days ago
Article
Open-source DeepResearch β Freeing our search agents
β’
636
Article
Ο0 and Ο0-FAST: Vision-Language-Action Models for General Robot Control
β’
67
reacted to
merve's
post with π
3 days ago
Post
3717
This week in open AI was π₯ Let's recap! π€
merve/january-31-releases-679a10669bd4030090c5de4d
LLMs π¬
> Huge: AllenAI released new TΓΌlu models that outperform DeepSeek R1 using Reinforcement Learning with Verifiable Reward (RLVR) based on Llama 3.1 405B π₯
> Mistral AI is back to open-source with their "small" 24B models (base & SFT), with Apache 2.0 license π±
> Alibaba Qwen released their 1M context length models Qwen2.5-Instruct-1M, great for agentic use with Apache 2.0 license π₯
> Arcee AI released Virtuoso-medium, 32.8B LLMs distilled from DeepSeek V3 with dataset of 5B+ tokens
> Velvet-14B is a new family of 14B Italian LLMs trained on 10T tokens in six languages
> OpenThinker-7B is fine-tuned version of Qwen2.5-7B-Instruct on OpenThoughts dataset
VLMs & vision π
> Alibaba Qwen is back with Qwen2.5VL, amazing new capabilities ranging from agentic computer use to zero-shot localization π₯
> NVIDIA released new series of Eagle2 models with 1B and 9B sizes
> DeepSeek released Janus-Pro, new any-to-any model (image-text generation from image-text input) with MIT license
> BEN2 is a new background removal model with MIT license!
Audio π£οΈ
> YuE is a new open-source music generation foundation model, lyrics-to-song generation
Codebase π©π»βπ»
> We are open-sourcing our SmolVLM training and eval codebase! https://github.com/huggingface/smollm/tree/main/vision
> Open-R1 is open-source reproduction of R1 by @huggingface science team https://huggingface.co/blog/open-r1
LLMs π¬
> Huge: AllenAI released new TΓΌlu models that outperform DeepSeek R1 using Reinforcement Learning with Verifiable Reward (RLVR) based on Llama 3.1 405B π₯
> Mistral AI is back to open-source with their "small" 24B models (base & SFT), with Apache 2.0 license π±
> Alibaba Qwen released their 1M context length models Qwen2.5-Instruct-1M, great for agentic use with Apache 2.0 license π₯
> Arcee AI released Virtuoso-medium, 32.8B LLMs distilled from DeepSeek V3 with dataset of 5B+ tokens
> Velvet-14B is a new family of 14B Italian LLMs trained on 10T tokens in six languages
> OpenThinker-7B is fine-tuned version of Qwen2.5-7B-Instruct on OpenThoughts dataset
VLMs & vision π
> Alibaba Qwen is back with Qwen2.5VL, amazing new capabilities ranging from agentic computer use to zero-shot localization π₯
> NVIDIA released new series of Eagle2 models with 1B and 9B sizes
> DeepSeek released Janus-Pro, new any-to-any model (image-text generation from image-text input) with MIT license
> BEN2 is a new background removal model with MIT license!
Audio π£οΈ
> YuE is a new open-source music generation foundation model, lyrics-to-song generation
Codebase π©π»βπ»
> We are open-sourcing our SmolVLM training and eval codebase! https://github.com/huggingface/smollm/tree/main/vision
> Open-R1 is open-source reproduction of R1 by @huggingface science team https://huggingface.co/blog/open-r1
- 1 reply
updated
a
Space
3 days ago
upvoted
an
article
4 days ago
Article
The AI tools for Art Newsletter - Issue 1
β’
44
posted
an
update
4 days ago
Post
2284
π R1 just built its own download dashboard!
Some fresh stats: +6M downloads for 800+ derivative models vs 2M for originals. Watch the numbers grow here: fdaudens/deepseek-download-stats
Some fresh stats: +6M downloads for 800+ derivative models vs 2M for originals. Watch the numbers grow here: fdaudens/deepseek-download-stats
upvoted
an
article
5 days ago
Article
Open-R1: Update #1
By
and 7 others
β’
β’
237
published
a
Space
7 days ago
posted
an
update
7 days ago
Post
3211
π― Kokoro TTS just hit v1.0! π
Small but mighty: 82M parameters, runs locally, speaks multiple languages. The best part? It's Apache 2.0 licensed!
This could unlock so many possibilities β¨
Check it out: hexgrad/Kokoro-82M
Small but mighty: 82M parameters, runs locally, speaks multiple languages. The best part? It's Apache 2.0 licensed!
This could unlock so many possibilities β¨
Check it out: hexgrad/Kokoro-82M
- 1 reply
reacted to
hexgrad's
post with π₯
7 days ago
Post
8212
hexgrad/Kokoro-82M got an upgrade! β¬οΈ More voices, more languages,
GitHub: https://github.com/hexgrad/kokoro
PyPI: https://pypi.org/project/kokoro/
Space: hexgrad/Kokoro-TTS
pip install kokoro, and still 82M parameters.GitHub: https://github.com/hexgrad/kokoro
PyPI: https://pypi.org/project/kokoro/
Space: hexgrad/Kokoro-TTS
- 11 replies
liked
a
model
8 days ago
liked
a
Space
8 days ago