Community Blog & Articles

Community Articles

How we OCR'ed 30,000 papers using Codex, open OCR models and Jobs

BidirLM: Turning Generative LLMs into the Best Open-Source Omnimodal Encoders

Using OCR models with llama.cpp

NEO-unify: Building Native Multimodal Unified Models End to End

"Darwin-27B-Opus: Surpassing the Foundation Model Without Training"

Uncensor any LLM with abliteration

Darwin V6: Diagnostic-Guided Evolutionary Model Merging

KV Caching Explained: Optimizing Transformer Inference Efficiency

Building Harvey-style tabular review from scratch, but better

When Speech AI Meets the Long Tail of Languages: Inside the VAANI Dataset

Stop benchmarking inference providers

about 18 hours ago

Mastering Tensor Dimensions in Transformers

Darwin-TTS: We Gave a TTS Model 3% of an LLM's Brain — It Started Showing Emotion

about 3 hours ago

How I contributed a new model to the Transformers library using Codex

A Review on the Evolvement of Load Balancing Strategy in MoE LLMs: Pitfalls and Lessons

How I Trained Action Chunking Transformer (ACT) on SO-101: My Journey, Gotchas, and Lessons

New in llama.cpp: Anthropic Messages API

LightOnOCR-2-1B: a lightweight high-performance end-to-end OCR model family

ArmBench-LLM 1.0: Benchmarking LLMs on Armenian Language Tasks

Releasing LiteCoder-Terminal-SFT

announcementdiffusionworld-model

Waypoint-1.5: Higher-Fidelity Interactive Worlds for Everyday GPUs

+1

multimodalnlpcommunity

Multimodal Embedding & Reranker Models with Sentence Transformers

ALTK‑Evolve: On‑the‑Job Learning for AI Agents

open-source-collabpartnershipsopen-source

Safetensors is Joining the PyTorch Foundation

multimodalon-devicegemma4

Welcome Gemma 4: Frontier multimodal intelligence on device

+3

Holo3: Breaking the Computer Use Frontier

Falcon Perception

gradioserveropen-source

Any Custom Frontend with Gradio's Backend

Granite 4.0 3B Vision: Compact Multimodal Intelligence for Enterprise Documents

Training mRNA Language Models Across 25 Species for $165

trlreinforcement-learningannouncement

TRL v1.0: Post-Training Library Built to Move with the Field

guideagentsinference-providers

Liberate your OpenClaw

+4

A New Framework for Evaluating Voice Agents (EVA)

Build a Domain-Specific Embedding Model in Under a Day

Community Articles

NEW Articles from Team or Enterprise organizations will get promoted to the main section.

How we OCR'ed 30,000 papers using Codex, open OCR models and Jobs

BidirLM: Turning Generative LLMs into the Best Open-Source Omnimodal Encoders

Using OCR models with llama.cpp

NEO-unify: Building Native Multimodal Unified Models End to End

"Darwin-27B-Opus: Surpassing the Foundation Model Without Training"

Uncensor any LLM with abliteration

Darwin V6: Diagnostic-Guided Evolutionary Model Merging

KV Caching Explained: Optimizing Transformer Inference Efficiency

Building Harvey-style tabular review from scratch, but better

When Speech AI Meets the Long Tail of Languages: Inside the VAANI Dataset

Stop benchmarking inference providers

about 18 hours ago

Mastering Tensor Dimensions in Transformers

Darwin-TTS: We Gave a TTS Model 3% of an LLM's Brain — It Started Showing Emotion

about 3 hours ago

How I contributed a new model to the Transformers library using Codex

A Review on the Evolvement of Load Balancing Strategy in MoE LLMs: Pitfalls and Lessons

How I Trained Action Chunking Transformer (ACT) on SO-101: My Journey, Gotchas, and Lessons

New in llama.cpp: Anthropic Messages API

LightOnOCR-2-1B: a lightweight high-performance end-to-end OCR model family

ArmBench-LLM 1.0: Benchmarking LLMs on Armenian Language Tasks

Releasing LiteCoder-Terminal-SFT

View all articles