DeepSeek-V3.1
Collection
3 items
•
Updated
•
220
Cris Almodovar PRO
calmodovar
AI & ML interests
NLP, log anomaly detection, cyber intelligence
Recent Activity
upvoted
a
collection
16 days ago
DeepSeek-V3.1
liked
a model
about 1 month ago
Tevatron/OmniEmbed-v0.1-multivent
Organizations
upvoted
a
collection
16 days ago
liked
a
model
about 1 month ago
reacted to
hba123's
post with 👀
about 2 months ago
Post
1372
As promised, and after the request of many, we have managed to fit in the first live session about Ark that we will be giving on the 28th of July.
For those who are already in the messaging channel, all is done, no need to do anything :-D For those interested in registering, please write to me at ark.robotics.uk@gmail.com - then I can add you and send you the invite.
We chose the timing to be 5 pm UK after consulting many of the interested people. Hope it works well for you too?
See you soon! Till then, have fun looking and using Ark: https://arkrobotics.notion.site/ARK-Home-22be053d9c6f8096bcdbefd6276aba61
pip install ark-roboticsFor those who are already in the messaging channel, all is done, no need to do anything :-D For those interested in registering, please write to me at ark.robotics.uk@gmail.com - then I can add you and send you the invite.
We chose the timing to be 5 pm UK after consulting many of the interested people. Hope it works well for you too?
See you soon! Till then, have fun looking and using Ark: https://arkrobotics.notion.site/ARK-Home-22be053d9c6f8096bcdbefd6276aba61
- 1 reply
liked
a
model
2 months ago
liked
2
models
3 months ago
reacted to
abidlabs's
post with 🔥
5 months ago
Post
3879
JOURNEY TO 1 MILLION DEVELOPERS
5 years ago, we launched Gradio as a simple Python library to let researchers at Stanford easily demo computer vision models with a web interface.
Today, Gradio is used by >1 million developers each month to build and share AI web apps. This includes some of the most popular open-source projects of all time, like Automatic1111, Fooocus, Oobabooga’s Text WebUI, Dall-E Mini, and LLaMA-Factory.
How did we get here? How did Gradio keep growing in the very crowded field of open-source Python libraries? I get this question a lot from folks who are building their own open-source libraries. This post distills some of the lessons that I have learned over the past few years:
1. Invest in good primitives, not high-level abstractions
2. Embed virality directly into your library
3. Focus on a (growing) niche
4. Your only roadmap should be rapid iteration
5. Maximize ways users can consume your library's outputs
1. Invest in good primitives, not high-level abstractions
When we first launched Gradio, we offered only one high-level class (gr.Interface), which created a complete web app from a single Python function. We quickly realized that developers wanted to create other kinds of apps (e.g. multi-step workflows, chatbots, streaming applications), but as we started listing out the apps users wanted to build, we realized what we needed to do:
Read the rest here: https://x.com/abidlabs/status/1907886
5 years ago, we launched Gradio as a simple Python library to let researchers at Stanford easily demo computer vision models with a web interface.
Today, Gradio is used by >1 million developers each month to build and share AI web apps. This includes some of the most popular open-source projects of all time, like Automatic1111, Fooocus, Oobabooga’s Text WebUI, Dall-E Mini, and LLaMA-Factory.
How did we get here? How did Gradio keep growing in the very crowded field of open-source Python libraries? I get this question a lot from folks who are building their own open-source libraries. This post distills some of the lessons that I have learned over the past few years:
1. Invest in good primitives, not high-level abstractions
2. Embed virality directly into your library
3. Focus on a (growing) niche
4. Your only roadmap should be rapid iteration
5. Maximize ways users can consume your library's outputs
1. Invest in good primitives, not high-level abstractions
When we first launched Gradio, we offered only one high-level class (gr.Interface), which created a complete web app from a single Python function. We quickly realized that developers wanted to create other kinds of apps (e.g. multi-step workflows, chatbots, streaming applications), but as we started listing out the apps users wanted to build, we realized what we needed to do:
Read the rest here: https://x.com/abidlabs/status/1907886
reacted to
fdaudens's
post with 🚀
5 months ago
Post
2393
See that purple banner on the Llama 4 models? It's Xet storage, and this is actually huge for anyone building with AI models. Let's geek out a little bit 🤓
Current problem: AI models are massive files using Git LFS. But with models getting bigger and downloads exploding, we needed something better.
Xet lets you version large files like code, with compression and deduplication, all Git-compatible. That means less bandwidth, faster sharing, and smoother collaboration.
Real numbers: ~25% deduplication on Llama 4 models, hitting ~40% for finetunes.
Scale matters here - the Hub served 2B model downloads in 30 days, Llama models alone at 60M. The upcoming Llama 4 Behemoth has 2T parameters! Xet's chunk-based system was built exactly for this.
This is the kind of engineering that makes the next wave of large models actually usable. Kudos to the team! 🧨
Check out the models collection: meta-llama/llama-4-67f0c30d9fe03840bc9d0164
Current problem: AI models are massive files using Git LFS. But with models getting bigger and downloads exploding, we needed something better.
Xet lets you version large files like code, with compression and deduplication, all Git-compatible. That means less bandwidth, faster sharing, and smoother collaboration.
Real numbers: ~25% deduplication on Llama 4 models, hitting ~40% for finetunes.
Scale matters here - the Hub served 2B model downloads in 30 days, Llama models alone at 60M. The upcoming Llama 4 Behemoth has 2T parameters! Xet's chunk-based system was built exactly for this.
This is the kind of engineering that makes the next wave of large models actually usable. Kudos to the team! 🧨
Check out the models collection: meta-llama/llama-4-67f0c30d9fe03840bc9d0164
reacted to
daavoo's
post with 🔥
5 months ago
Post
1838
After working on agent evaluation🔍🤖 the last weeks, we started to accumulate code to make trying different agent frameworks easier. From that code, we have built and just released a small library called
Give it a try and a ⭐: https://github.com/mozilla-ai/any-agent
any-agent. Give it a try and a ⭐: https://github.com/mozilla-ai/any-agent
from any_agent import AgentConfig, AgentFramework, AnyAgent
agent = AnyAgent.create(
AgentFramework("smolagents"), # or openai, langchain, llama_index
AgentConfig(
model_id="gpt-4o-mini"
)
)
agent.run("Which Agent Framework is the best??")liked
a
model
6 months ago
reacted to
openfree's
post with 🚀
7 months ago
Post
4388
🌟 MoneyRadar - AI-Powered Global News Analysis System
💻 Live Demo: openfree/MoneyRadar
🎯 Core Features
1. 🤖 24/7 Automated News Scanning
Auto-collection of Top 100 trending news
Real-time monitoring across 60 countries
Smart filtering of investment-critical news
2. 🔍 Advanced Custom Search
Unlimited keyword search capability
Country/language-specific search options
Real-time trend-based related keywords
3. 🎨 Smart Analysis & Visualization
AI-powered sentiment analysis
Automated content summarization
Investment decision-supporting insights
⚡ Automated Information Collection
Key Companies (NVIDIA, APPLE, TESLA, etc.)
Earnings/Forecasts
Product/Technology announcements
Market share changes
M&A and major news
Financial Markets & Digital Assets
Macroeconomic indicators
Regulatory changes
Market sentiment analysis
Major exchange updates
📊 Business Applications
Real-time market trend tracking
Competitor movement monitoring
Early investment opportunity detection
Risk early warning system
🌟 Key Differentiators
Full Automation
Zero manual intervention
Real-time data updates
Automated result storage/management
User-Centric Design
Intuitive interface
Customizable alerts
Mobile optimization
Advanced Analytics
News cross-checking
Historical tracking
Trend prediction support
Join Community 💬
"With MoneyRadar, never miss a beat in the global market movements!"
💻 Live Demo: openfree/MoneyRadar
🎯 Core Features
1. 🤖 24/7 Automated News Scanning
Auto-collection of Top 100 trending news
Real-time monitoring across 60 countries
Smart filtering of investment-critical news
2. 🔍 Advanced Custom Search
Unlimited keyword search capability
Country/language-specific search options
Real-time trend-based related keywords
3. 🎨 Smart Analysis & Visualization
AI-powered sentiment analysis
Automated content summarization
Investment decision-supporting insights
⚡ Automated Information Collection
Key Companies (NVIDIA, APPLE, TESLA, etc.)
Earnings/Forecasts
Product/Technology announcements
Market share changes
M&A and major news
Financial Markets & Digital Assets
Macroeconomic indicators
Regulatory changes
Market sentiment analysis
Major exchange updates
📊 Business Applications
Real-time market trend tracking
Competitor movement monitoring
Early investment opportunity detection
Risk early warning system
🌟 Key Differentiators
Full Automation
Zero manual intervention
Real-time data updates
Automated result storage/management
User-Centric Design
Intuitive interface
Customizable alerts
Mobile optimization
Advanced Analytics
News cross-checking
Historical tracking
Trend prediction support
Join Community 💬
"With MoneyRadar, never miss a beat in the global market movements!"
reacted to
merve's
post with 🔥
7 months ago
Post
5508
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
upvoted
a
collection
7 months ago
reacted to
merve's
post with 🔥
8 months ago
Post
4940
supercharge your LLM apps with smolagents 🔥
however cool your LLM is, without being agentic it can only go so far
enter smolagents: a new agent library by Hugging Face to make the LLM write code, do analysis and automate boring stuff!
Here's our blog for you to get started https://huggingface.co/blog/smolagents
however cool your LLM is, without being agentic it can only go so far
enter smolagents: a new agent library by Hugging Face to make the LLM write code, do analysis and automate boring stuff!
Here's our blog for you to get started https://huggingface.co/blog/smolagents
reacted to
MoritzLaurer's
post with 👍
9 months ago
Post
2639
Quite excited by the ModernBERT release! 0.15/0.4B small, 2T modern pre-training data and tokenizer with code, 8k context window, great efficient model for embeddings & classification!
This will probably be the basis for many future SOTA encoders! And I can finally stop using DeBERTav3 from 2021 :D
Congrats @answerdotai , @LightOnIO and collaborators like @tomaarsen !
Paper and models here 👇https://huggingface.co/collections/answerdotai/modernbert-67627ad707a4acbf33c41deb
This will probably be the basis for many future SOTA encoders! And I can finally stop using DeBERTav3 from 2021 :D
Congrats @answerdotai , @LightOnIO and collaborators like @tomaarsen !
Paper and models here 👇https://huggingface.co/collections/answerdotai/modernbert-67627ad707a4acbf33c41deb
- 3 replies
upvoted
an
article
10 months ago
Article
Visually Multilingual: Introducing mcdse-2b
By
•
•
41
reacted to
singhsidhukuldeep's
post with 👀
11 months ago
Post
2167
While Google's Transformer might have introduced "Attention is all you need," Microsoft and Tsinghua University are here with the DIFF Transformer, stating, "Sparse-Attention is all you need."
The DIFF Transformer outperforms traditional Transformers in scaling properties, requiring only about 65% of the model size or training tokens to achieve comparable performance.
The secret sauce? A differential attention mechanism that amplifies focus on relevant context while canceling out noise, leading to sparser and more effective attention patterns.
How?
- It uses two separate softmax attention maps and subtracts them.
- It employs a learnable scalar λ for balancing the attention maps.
- It implements GroupNorm for each attention head independently.
- It is compatible with FlashAttention for efficient computation.
What do you get?
- Superior long-context modeling (up to 64K tokens).
- Enhanced key information retrieval.
- Reduced hallucination in question-answering and summarization tasks.
- More robust in-context learning, less affected by prompt order.
- Mitigation of activation outliers, opening doors for efficient quantization.
Extensive experiments show DIFF Transformer's advantages across various tasks and model sizes, from 830M to 13.1B parameters.
This innovative architecture could be a game-changer for the next generation of LLMs. What are your thoughts on DIFF Transformer's potential impact?
The DIFF Transformer outperforms traditional Transformers in scaling properties, requiring only about 65% of the model size or training tokens to achieve comparable performance.
The secret sauce? A differential attention mechanism that amplifies focus on relevant context while canceling out noise, leading to sparser and more effective attention patterns.
How?
- It uses two separate softmax attention maps and subtracts them.
- It employs a learnable scalar λ for balancing the attention maps.
- It implements GroupNorm for each attention head independently.
- It is compatible with FlashAttention for efficient computation.
What do you get?
- Superior long-context modeling (up to 64K tokens).
- Enhanced key information retrieval.
- Reduced hallucination in question-answering and summarization tasks.
- More robust in-context learning, less affected by prompt order.
- Mitigation of activation outliers, opening doors for efficient quantization.
Extensive experiments show DIFF Transformer's advantages across various tasks and model sizes, from 830M to 13.1B parameters.
This innovative architecture could be a game-changer for the next generation of LLMs. What are your thoughts on DIFF Transformer's potential impact?
- 1 reply
upvoted
a
collection
about 1 year ago
upvoted
an
article
about 1 year ago
Article
Fine-tune Llama 3.1 Ultra-Efficiently with Unsloth
By
•
•
356
upvoted
an
article
over 1 year ago
Article
Multimodal Augmentation for Documents: Recovering “Comprehension” in “Reading and Comprehension” task
By
•
•
17