AI news & updates

OTIS Mock AIME 2024-2025 99.2%

FrontierMath-Tiers-1-3-v2-Private 74.4% · FrontierMath-Tier-4-v2-Private 41.5% · Chess Puzzles 35.0% · Mystery Game Puzzles 14.1%

Schematron V2 Turbo is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes throughput for high-volume extraction workloads. Extraction instructions must be supplied through a JSON schema in re

Schematron V2 Small is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes extraction quality for complex schemas and long pages. Extraction instructions must be supplied through a JSON schema

Fugu Max is the cost-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to route...

Fugu Ultra v2 is the higher-performance model in Sakana AI's Fugu family. Rather than a single monolithic model, Fugu is a learned multi-agent orchestration system: a language model trained to...

Furniture Assembly 34.2%

Furniture Assembly 21.7%

Ling 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding native visual perception and advanced visual...

MirrorCode 73.3% · Furniture Assembly 70.0%

Furniture Assembly 20.0%

Furniture Assembly 26.7%

Furniture Assembly 23.3%

Furniture Assembly 31.7%

Furniture Assembly 26.7%

Furniture Assembly 35.8%

Furniture Assembly 60.8%

Furniture Assembly 42.5%

Furniture Assembly 33.3%

Furniture Assembly 28.3%

Furniture Assembly 28.3%

Furniture Assembly 42.5%

Furniture Assembly 80.0%

Furniture Assembly 56.7%

Furniture Assembly 54.2%

Furniture Assembly 44.2%

Furniture Assembly 38.3%

Furniture Assembly 37.5%

Mercury 2.5 is the fastest reasoning LLM, and the latest diffusion LLM (dLLM) from Inception. Instead of generating tokens sequentially, Mercury 2.5 produces and refines multiple tokens in parallel, achieving...

EBR-bench 57.1%

Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...

Nex-N2.5 is an agentic model built to turn goals into working, verified outcomes. Its core strength is agentic coding within a visual feedback loop: it can explore codebases, implement multi-file...

EBR-bench 45.7%