AI Research

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

1153 articles · page 37

Latest

Google's Gemma 3n Crams 5B-Parameter Brains Into Just 2GB of VRAM

SGLang taps Hugging Face transformers as a backend, instantly unlocking models like Kyutai's Helium

Fine-tune FLUX.1-dev on an RTX 4090: QLoRA slashes VRAM below 10GB

Groq's LPU chips land on Hugging Face, bringing sub-100ms inference to open models

vLLM's long prompts silently choke fast token generation for everyone

Hugging Face adds Featherless AI for serverless access to models like DeepSeek-R1

Hugging Face Kernel Hub erases 96 GB and hours of build pain with a single import

NVIDIA's GR00T N1.5 robot model now fine-tunes on a $200 open-source arm

Hugging Face and NVIDIA team up to rent GPU clusters by the training run

Hugging Face's ScreenSuite stress-tests 5 AI models on pure vision — no DOM cheating

KV Caching from Scratch in nanoVLM Delivers a 38% Generation Speedup

Stability AI’s audio model runs on Arm CPUs, generating WAVs in seconds

H Company drops open-source Holo1 VLMs that slash web automation costs to $0.13 per task

SmolVLA packs 30% faster robot learning into a 450M model you can run on a MacBook

TRL v0.18.0 lets you train and serve LLMs on the same GPUs, cutting idle time

JSON plus Python beats raw code: structured agents gain 7 points on reasoning benchmarks

DeepSpeed ZeRO-3 clashing with Liger GRPO loss? Shape mismatch halts Qwen2.5 finetuning

Dell ships Llama 4 in under an hour, adds full on-prem AI app catalog

Hugging Face's 'tiny-agents' Builds a Python AI Agent in Just 70 Lines of Code

Falcon-H1 Packs 7B-Class Punch Into a 0.5B Model Using a Hybrid Attention-SSM Brain

Falcon-Arabic 7B beats models 4x its size on Arabic benchmarks

Hugging Face drops nanoVLM: a 135M-param vision model you can train from scratch in pure PyTorch

New Diffusers quantization slashes Flux GPU memory from 31GB to under 18GB

10,000+ Hugging Face models hit Azure AI Foundry with same-day release promise