AI Pulse by Inblix

AI Research

Cutting-edge AI research papers, breakthroughs, and academic developments.

595 articles

Hugging Face tries to clone OpenAI’s Deep Research in 24 hours with open code agent — Inblix summary
Research

Hugging Face tries to clone OpenAI’s Deep Research in 24 hours with open code agent

Hugging Face Blog · Feb 4, 2025 · 2 min read

OpenAI’s new Deep Research tool is undeniably slick—it browses the web, synthesizes multi-step answers, and just scored...

Physical Intelligence drops π0-FAST: a VLA model controlling 7 robots on 68 tasks — Inblix summary
Research

Physical Intelligence drops π0-FAST: a VLA model controlling 7 robots on 68 tasks

Hugging Face Blog · Feb 4, 2025 · 2 min read

Physical Intelligence just open-sourced π0 and π0-FAST, a pair of vision-language-action models that mark a serious ste...

DeepSeek R1's 20,000-token responses are breaking GPUs and rewriting the rules of evaluation — Inblix summary
Research

DeepSeek R1's 20,000-token responses are breaking GPUs and rewriting the rules of evaluation

Hugging Face Blog · Feb 2, 2025 · 3 min read

The open-source race to replicate DeepSeek R1 just hit its first major wall: the model's runaway verbosity. The Open-R1...

How a 3B open model taught itself to rethink math—no human data needed — Inblix summary
Research

How a 3B open model taught itself to rethink math—no human data needed

Hugging Face Blog · Jan 31, 2025 · 2 min read

The release of DeepSeek-R1 sent a jolt through the AI world. Here was an open model going toe-to-toe with OpenAI's o1 o...

Flux.1 beat Midjourney and DALL-E 3 in 2024 — here's the open-source shift it triggered — Inblix summary
Research

Flux.1 beat Midjourney and DALL-E 3 in 2024 — here's the open-source shift it triggered

Hugging Face Blog · Jan 31, 2025 · 2 min read

It’s almost jarring to remember that just a year ago, we were all laughing at AI’s inability to count fingers. The pace...

DeepSeek R1 deployment on AWS now costs just $8.30/hour via Hugging Face endpoints — Inblix summary
Research

DeepSeek R1 deployment on AWS now costs just $8.30/hour via Hugging Face endpoints

Hugging Face Blog · Jan 30, 2025 · 2 min read

DeepSeek dropped a bomb on the AI world last week with R1, an open-source reasoning model that goes toe-to-toe with Ope...

DeepSeek-R1 cracked open: New Open-R1 project reverse-engineers the secret RL recipe — Inblix summary
Research

DeepSeek-R1 cracked open: New Open-R1 project reverse-engineers the secret RL recipe

Hugging Face Blog · Jan 28, 2025 · 2 min read

DeepSeek's R1 model didn't just match OpenAI's o1 on reasoning benchmarks—it tanked Nvidia's stock and reshuffled the A...

Hugging Face unifies serverless inference with SambaNova, Replicate, and more — Inblix summary
Research

Hugging Face unifies serverless inference with SambaNova, Replicate, and more

Hugging Face Blog · Jan 28, 2025 · 2 min read

Hugging Face is finally opening up its serverless Inference API to third-party providers, a move that acknowledges a si...

Open video models still can't touch image AI's 'Stable Diffusion moment' — Inblix summary
Research

Open video models still can't touch image AI's 'Stable Diffusion moment'

Hugging Face Blog · Jan 27, 2025 · 2 min read

If you've been waiting for open-source video generation to have its Stable Diffusion moment, keep waiting. A new deep-d...

Hugging Face brings sight to smolagents, unlocking visual web browsing for AI — Inblix summary
Research

Hugging Face brings sight to smolagents, unlocking visual web browsing for AI

Hugging Face Blog · Jan 24, 2025 · 2 min read

Hugging Face just tore down a major wall for AI agents. Their lightweight agent framework, smolagents, can now see, whi...

NVIDIA’s KVPress toolkit slashes 1M-token Llama 3 memory from 330GB to fit on a single GPU — Inblix summary
Research

NVIDIA’s KVPress toolkit slashes 1M-token Llama 3 memory from 330GB to fit on a single GPU

Hugging Face Blog · Jan 23, 2025 · 3 min read

If you’ve ever tried to run a model with a million-token context window, you already know the math is brutal. For Llama...

The world’s smallest VLM is here at 256M parameters, beating 80B models from 17 months ago — Inblix summary
Research

The world’s smallest VLM is here at 256M parameters, beating 80B models from 17 months ago

Hugging Face Blog · Jan 23, 2025 · 2 min read

Hugging Face just shrank its vision language models down to sizes that would have sounded like a joke a year and a half...

Hugging Face taps FriendliAI, ranked fastest GPU inference, for 1-click H100 deployment — Inblix summary
Research

Hugging Face taps FriendliAI, ranked fastest GPU inference, for 1-click H100 deployment

Hugging Face Blog · Jan 22, 2025 · 2 min read

Hugging Face just gave its users a faster on-ramp to production AI. A new partnership with FriendliAI—ranked by Artific...

Hugging Face finally lets orgs publish blog posts—but there's a paywall — Inblix summary
Research

Hugging Face finally lets orgs publish blog posts—but there's a paywall

Hugging Face Blog · Jan 20, 2025 · 2 min read

Hugging Face shipped a feature on January 20, 2025 that seems so basic you'd assume it already existed: organizations c...

Hugging Face bridges its entire pipeline to timm's 200K daily users via TimmWrapper — Inblix summary
Research

Hugging Face bridges its entire pipeline to timm's 200K daily users via TimmWrapper

Hugging Face Blog · Jan 16, 2025 · 2 min read

The wall between two of PyTorch's most popular libraries just came down. A new integration called TimmWrapper lets you...

Hugging Face's TGI adds vLLM and TRT-LLM backends to end the serving wars — Inblix summary
Research

Hugging Face's TGI adds vLLM and TRT-LLM backends to end the serving wars

Hugging Face Blog · Jan 16, 2025 · 2 min read

Hugging Face is finally addressing the fragmentation that has turned LLM serving into a choose-your-own-headache exerci...

Hugging Face’s static embeddings hit 400x CPU speedup while keeping 85% accuracy — Inblix summary
Research

Hugging Face’s static embeddings hit 400x CPU speedup while keeping 85% accuracy

Hugging Face Blog · Jan 15, 2025 · 2 min read

Hugging Face just dropped a pair of embedding models that flip the performance-to-efficiency tradeoff on its head. The...

AI autonomy is a loaded gun: Google warns fully autonomous agents risk total loss of human control — Inblix summary
Research

AI autonomy is a loaded gun: Google warns fully autonomous agents risk total loss of human control

Hugging Face Blog · Jan 13, 2025 · 3 min read

The conversation around AI agents is shifting from "what can they do" to "what should we let them do," and a new analys...

VDR-2B-Multi obliterates OCR pipelines with 500K-sample multilingual visual search — Inblix summary
Research

VDR-2B-Multi obliterates OCR pipelines with 500K-sample multilingual visual search

Hugging Face Blog · Jan 10, 2025 · 2 min read

Forget OCR and chunking strategies. The new open-source vdr-2b-multi-v1 model from LlamaIndex and MrLight lets you sear...

Community fine-tunes are crushing official models on carbon efficiency, Open LLM Leaderboard data reveals — Inblix summary
Research

Community fine-tunes are crushing official models on carbon efficiency, Open LLM Leaderboard data reveals

Hugging Face Blog · Jan 9, 2025 · 2 min read

The Open LLM Leaderboard just got a lot more interesting—and a little greener. The team behind the popular benchmarking...

Hugging Face's smolagents makes code-writing AI agents dirt simple in 3 lines — Inblix summary
Research

Hugging Face's smolagents makes code-writing AI agents dirt simple in 3 lines

Hugging Face Blog · Dec 31, 2024 · 2 min read

Hugging Face just dropped `smolagents`, a new library that strips building AI agents down to its absolute bones. We're...

PyTorch's Hidden Memory Hog: Why Your 200MB Tensor Needs 3GB During Training — Inblix summary
Research

PyTorch's Hidden Memory Hog: Why Your 200MB Tensor Needs 3GB During Training

Hugging Face Blog · Dec 24, 2024 · 2 min read

That `CUDA out of memory` error isn't just frustrating—it's a lie of omission. It tells you the GPU is full, but not wh...

NVIDIA's LogitsProcessorZoo Lets You Rewrite an LLM's Brain in Real Time — Inblix summary
Research

NVIDIA's LogitsProcessorZoo Lets You Rewrite an LLM's Brain in Real Time

Hugging Face Blog · Dec 23, 2024 · 2 min read

Everyone talks about prompt engineering, but that's like trying to steer a car by shouting directions from the backseat...

GPT-4o's Reasoning Plummets 26 Points When Listening Instead of Reading — Inblix summary
Research

GPT-4o's Reasoning Plummets 26 Points When Listening Instead of Reading

Hugging Face Blog · Dec 20, 2024 · 2 min read

Give a state-of-the-art model a logic puzzle in writing, and it aces it. Read the same puzzle aloud, and that performan...