Technology ❯ Artificial Intelligence ❯ Model Evaluation

Benchmarking

Performance Metrics Performance Comparison User Feedback Performance Assessment Human Evaluation Reasoning Performance AIME and MATH Tests IFEval Benchmark Experimental Methodology UAMO Scores

Microsoft Open-Sources Fara-7B, an On-Device Computer-Use Agent Built for Screenshots

The research preview interprets UIs from pixels to issue grounded actions for private, low-latency automation on consumer hardware.

xAI Rolls Out Grok 4.1 Free to All Users, Citing Faster, More Accurate Replies

New Papers Flag Retrieval as RAG’s Weak Link, Propose Practical Fixes

Samsung’s Tiny Recursive Model Tops Bigger AIs on Tough Reasoning Benchmarks

Meta Releases Open-Weights Code World Model With Built-In Execution Simulation

DeepSeek Releases V3.1‑Terminus With Reliability Fixes, Mode Limits and Agent Gains

OpenAI Says Incentives Drive AI Hallucinations, Calls for Scoreboard Overhaul

Google Launches Gemma 3 270M to Enable Hyper-Efficient On-Device AI

Apple Study Finds Reasoning Models Collapse on Complex Puzzles

Microsoft Unveils Phi-4 Reasoning Models That Outperform Larger AI Systems

Meta Faces Backlash Over Use of Experimental AI Model for Benchmark Testing

Mistral Unveils Small 3 AI Model, Rivaling Larger Competitors in Efficiency and Accuracy

Alibaba Launches QwQ-32B AI Model to Challenge OpenAI's Reasoning Models

Mistral Unveils Pixtral 12B, Its First Multimodal AI Model