AI Wiki — Knowledge Base

CrofAI 'cheapest inference provider in the world' gets exposed as an OpenRouter wrapper, routing requests to smaller, cheaper models at up to 20x markup. CrofAI responds to Wire Fraud allegations by denying everything, then backtracking, then 3 hours later wiping their entire online presence

Disclaimer: no AI was used whatsoever to write this post Cautionary tale about chasing cheap tokens. exposé: https://kendell.dev/blog/crofaifalse/ reaction by nahcrof, announcing the shutdown of the s

model-releasesr-localllama

15 Sept 2026

Search: “models”

Sources: the US restores the UK AISI's access to the latest US models for safety testing, after ordering OpenAI and Anthropic to stop sharing models last month (Tony Diver/Telegraph)

Just another purely open-weight models benchmark

Beyond the Leaderboard: Multi-Dimensional Evaluation of Dense and Mixture-of-Experts Models for Automated Program Repair

Reading, Not Manipulating: Leveraging Router Logits for Multimodal Safety in MoE Vision-Language Models

zkLLMPoT: Efficient Zero Knowledge Proof of Training for Large Language Models

Adaptive Mutual Distillation for Balanced Multi-Task Post-Training of Large Language Models

Overcoming Challenges of Interpretive Structural Modeling with Large Language Models

Unmasking Propaganda: A Comparative Analysis of Masked and Causal Language Models

On the Divergence of Accuracy and Mechanism Consistency in Time Series World Models

Comparison of techniques for fine-tuning open-weight models for entity extraction from radiology reports

DrivingBench: Can Vision-Language Models Drive a Toyota Corolla?

NinaXander: Feasibility and Limits of Composing Frozen Language Models Across Architecture Families via a Shared Latent Space

When a Kindergartener Solves Calculus: Measuring Capability Leakage in Role-Prompted Reasoning Models

A Survey on Fake Review Detection: From Pre-trained Language Models to Large Language Models

Models Got Talent: Identifying High Performing Wearable Human Activity Recognition Models Without Training

Visual Representation and History Modeling for Navigation World Models

Causal Localization of the Refusal Direction in Audio Language Models

Few-Shot Degradation Is Not What It Seems: Behavioral Evidence, Representation Analysis, and a Random-Text Control Across 12 Models, 2 Tasks, and 2 Architectures

Self-reported archetypes and behavioral failures in Large Language Models

A Controlled Reevaluation of Coreference Resolution Models

Biomedical Reference Generation Remains Unreliable across 26 Large Language Models

CrofAI 'cheapest inference provider in the world' gets exposed as an OpenRouter wrapper, routing requests to smaller, cheaper models at up to 20x markup. CrofAI responds to Wire Fraud allegations by denying everything, then backtracking, then 3 hours later wiping their entire online presence

RxScribe Bench: A Multi-Axis Benchmark for Evaluating Vision-Language Models on Indian Outpatient Prescriptions

Internal OpenAI docs detail contractors evaluating anonymized prompts and chats to improve the models; model training is turned on by default for consumer plans (Joseph Cox/404 Media)

Diffs vs. Whole Files: An Empirical Comparison of Iterative Edit-Based and Direct Generation for Flutter/Dart Code Models

Learning to Use Imagination: Progress-Conditioned Future Utilization for World Action Models

Govern models with MLflow and Amazon SageMaker AI Model Registry sync: Part 1

Which local models handle structured extraction well? I moved my agent's memory into a folder (BYO model) and want to run it on Ollama instead of Sonnet

Open models are closing the intelligence gap, says @ollama's @jmorgan. That opens up a new opportunity for founders: coordinating fast, chea…

One Model, Many Graphs: Learning over Attributed Graphs across Heterogeneous Modalities with Vision-Language Models

WinoQueer-NL: Assessing Bias in Dutch Language Models toward LGBTQ+ Identities

All currently popular local models in one table + Opus 4.8 results

Chain-of-Thought Faithfulness of Reasoning Models Varies with Where and How Preference Cues Are Delivered

Evaluating Tiny Recursive Models Across Training for Code Generation

Integrating adaptive human behavior into epidemic models with large language models

Sleight of Word Benchmark: Can Language Models Notice If Their Own Output Was Tampered With?

Small Reward Models via Backward Inference

Successive Capacity Growth: Task-Complexity-Driven Width and Depth Expansion for Vision Transformer Encoders in JEPA World Models

LeFlow: Generative Latent Flow Planning for World Models

Improved denoising diffusion probabilistic models with efficient non-diagonal covariance modeling

On the Capability Separation Between World-Model Policy Learning and Imitated World-Action Models

Researchers detail the growing use of AI in cyberattacks across many Chinese state-linked groups, primarily using open-weight models like Kimi K3 and DeepSeek (Mark Anderson/Bloomberg)

1-bit / 2-bit / Ternary / Bitnet Models - Updates & Tracking

Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search

Jais 2: A Family of Arabic-Centric Open Large Language Models

Interesting new research from IBM. If you pick models from benchmark deltas, some of that delta belongs to the phrasing rather than the mode…

Beyond Static Models: An Evolving Framework for Continual Learning in Large Language Models across Training Stages

The small open weight models are scarier in AI development