Can you train a model on Simon Willison's deeply unscientific pelican benchmark?
sergiopaniego
• • 2NEW Articles from Team or Enterprise organizations will get promoted to the main section.
What Do Memory Benchmarks Actually Measure? (Hint: Not Storage)
mjfk
• • 1Geometric Memory FT4 — Distill Against a Consensus, Ship a Rotation
AbstractPhil
• • 1I Built a RAG System, Then Tried to Break It
nazeerbashashaik
• • 1🎲 Apprendre à un réseau à écrire avec seulement une récompense — il a atteint 99,9 % grammatical sans apprendre une seule règle 🇫🇷
RDTvlokip
• • 5🎲 Teaching a network to write with reward only — it hit 99.9% grammatical without learning a single rule 🇫🇷
RDTvlokip
• • 124/24 Retrieval, Yet the Embeddings Changed: An MLX Q4–Q8 Sweep with CUDA Controls
TiGa-RCE
• • 2LettucePrevent - Real-Time Prevention of Factual Hallucinations in RAG
lebe1
• • 9Training a 2.7B MoE from scratch for $200, one GPU at a time
vovaRL
• • 5VisionPsy-Nano: State-of-the-Art On-Device Vision-Language Models
qvac
• • 15Bekko Embedding: how small can a multilingual retrieval model be?
hotchpotch
• • 6Wan 3.0 Ecosystem Signals: What WanSong, Wan-Dancer, and Wan-Streamer Reveal About Alibaba's Next Video Model
ResterChed
• • 2Building a Bloom Filter for 1.4 Billion Web Captures
brian-learns
• A Week of Releases from Consumer GPUs
spoodyzz
• The OlmoEarth Platform: Geospatial inference at planetary scale
allenai
• • 40ECMWF's AI forecasting model is open source: now let's make it easy to run.
hugging-science
• • 11LFM2.5-Encoders for Fast Long-Context Inference on CPU
LiquidAI
• • 68The Kimi K3 Receipt
KissTheHabit
• Skyvern + CapSolver: a controlled browser-automation pattern
CapSolverAI
• How Gateway Uses Hugging Face to Store and Train Its Freight AI Model
gatewaylines
• • 1