wazn Collection Jev-like decision models, trained using custom architecture and training recipe β’ 1 item β’ Updated 23 days ago
SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers Paper β’ 2609.01343 β’ Published Sep 1 β’ 63
view article Article Beyond LoRA: Can you beat the most popular fine-tuning technique? +2 BenjaminB, sayakpaul, hubnemo, kashif β’ Jun 18 β’ 110
Running 19 Defeating the trainer-generator precision mismatch in TRL π― 19 Download research PDF (Pro access required)
view article Article β‘ nano-vLLM: Lightweight, Low-Latency LLM Inference from Scratch zamal β’ Jun 28, 2025 β’ 50
jinaai/jina-embeddings-v5-text-small-text-matching Sentence Similarity β’ 0.6B β’ Updated Apr 15 β’ 3.89k β’ 12
view article Article Finally, a Replacement for BERT: Introducing ModernBERT +13 bwarner, NohTow, bclavie, orionweller, ohallstrom, staghado, alexisgallagher, rbiswasfc, fladhak, tomaarsen, ncoop57, griffin, jph00, johnowhitaker, iacolippo β’ Dec 19, 2024 β’ 757
view article Article Accelerate ND-Parallel: A guide to Efficient Multi-GPU Training +3 smohammadi, siro1, winglian, marcsun13, djsaunde β’ Aug 8, 2025 β’ 102
Running on CPU Upgrade Featured 3.39k The Smol Training Playbook π 3.39k The secrets to building world-class LLMs
LangAGI-Lab/magpie-reasoning-v1-100k-math-verifiable Viewer β’ Updated Feb 9, 2025 β’ 21.7k β’ 65 β’ 1
Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning Paper β’ 2509.24372 β’ Published Sep 29, 2025 β’ 16
Running 123 Unlocking On-Policy Distillation for Any Model Family π 123 Explore onβpolicy distillation visualizations for any model