mDenseOn with the mLateOn: Open Multilingual, Long-Context, and Code Retrieval Models
lightonai
• • 33Community Blog & Articles
Community Articles
view all
announcementmlxllm
The PR you would have opened yourself
72
April 16, 2026 
multimodalnlpcommunity
Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers
78
April 16, 2026 
Inside VAKRA: Reasoning, Tool Use, and Failure Modes of Agents
28
April 15, 2026 Meet HoloTab by HCompany. Your AI browser companion.
24
April 15, 2026 
announcementdiffusionworld-model
Waypoint-1.5: Higher-Fidelity Interactive Worlds for Everyday GPUs
- +1
34
April 9, 2026 multimodalnlpcommunity
Multimodal Embedding & Reranker Models with Sentence Transformers
69
April 9, 2026 
open-source-collabpartnershipsopen-source
Safetensors is Joining the PyTorch Foundation
41
April 8, 2026 
multimodalon-devicegemma4
HotWelcome Gemma 4: Frontier multimodal intelligence on device
- +3
919
April 2, 2026 Falcon Perception
70
April 1, 2026 
gradioserveropen-source
Any Custom Frontend with Gradio's Backend
38
April 1, 2026 Granite 4.0 3B Vision: Compact Multimodal Intelligence for Enterprise Documents
34
March 31, 2026 
Training mRNA Language Models Across 25 Species for $165
28
March 31, 2026 
trlreinforcement-learningannouncement
TRL v1.0: Post-Training Library Built to Move with the Field
58
March 31, 2026 
guideagentsinference-providers
Liberate your OpenClaw
- +4
49
March 27, 2026 NEW Articles from Team or Enterprise organizations will get promoted to the main section.
mDenseOn with the mLateOn: Open Multilingual, Long-Context, and Code Retrieval Models
lightonai
• • 33The Fast Gemma Challenge: our verified-SOTA recipe, in full
FINAL-Bench
• • 22LFM2.5-Encoders for Fast Long-Context Inference on CPU
LiquidAI
• • 65Accelerating Qwen3.6 on Intel® Core™ Ultra Series 3 with DFlash
ofirzaf
• • 11Kimi K3 Model Overview: 2.8T Parameters, MXFP4 Quantization, and What the Open Weights Mean for the Community
ResterChed
• • 194Training a coding agent using the OpenCode harness in remote HF sandboxes with TRL and OpenEnv
sergiopaniego
• • 9超越表层对齐：信念是通往深层对齐的新入口
Alibaba-VELLDEPTH
• • 9Code a simple RAG from scratch
ngxson
• • 375VisionPsy-Nano: State-of-the-Art On-Device Vision-Language Models
qvac
• • 15Bekko Embedding: how small can a multilingual retrieval model be?
hotchpotch
• • 5Language Identification for 42 Indian Languages: Opening New Frontiers in Indic Speech AI
ARTPARK-IISc
• • 5LeRobot Humanoid: An Open, Low-Cost, 3D-Printed Humanoid for Robot Learning
VirgileBatto
• • 70FLUX 3 Model Overview: Multimodal Flow Models for Image, Video, Audio, and Action Prediction
ResterChed
• • 12LettucePrevent - Real-Time Prevention of Factual Hallucinations in RAG
lebe1
• • 9Training a 2.7B MoE from scratch for $200, one GPU at a time
vovaRL
• • 5jina-reranker-v3.5: Faster Listwise Reranking with Hybrid Attention and Self-Distillation
jinaai
• • 4Uncensor any LLM with abliteration
mlabonne
• • 886KV Caching Explained: Optimizing Transformer Inference Efficiency
not-lain
• • 388NEO-unify: Building Native Multimodal Unified Models End to End
sensenova
• • 174Hugging Face on AMD Instinct MI455X: First Transformers Results
badaoui
• • 18