Benchmarks

Define the Frontier and Track the Acceleration.

GPT-6-Astra / Full 13.98% 路 Scientific Reproduction

PaperBenchX

93 paper-reproduction tasks across 12 research areas, scored through clean replay and evidence-backed scientific rubrics.

馃弳 Pine Computer / PCR 路 CS 78.3% 路 Computer-Use Agents

SaaS-Bench v1.1

v1.1 results across 106 tasks and 23 SaaS systems, comparing generic and native harnesses using Checkpoint Score and Resolved Score.

馃弳 2026.06: GPT-5.5-High 路 Coding

Monthly-SWEBench

A continuously updated benchmark evaluating AI coding agents on real-world software engineering tasks from GitHub issues.

馃弳 Claude-Opus-4.7: 39.1% 路 Coding

RoadmapBench

115 real-world version-upgrade tasks from 17 repositories across 5 programming languages. Evaluates whether coding agents can implement substantial new functionality guided only by a multi-target specification.

Opus-4.8: 59.1 Dataset Score 路 Coding

EvoCode-Bench

26 multi-turn coding tasks with 227 evaluated rounds. Each task keeps the same workspace and agent session while requirements change, accumulate, and sometimes conflict.

Claude Opus 4.7: 44.6% V2 Reward 路 Computer-Use Agents

ClawBench

Can AI browser agents complete everyday online tasks on live websites? ClawBench evaluates agents across V1/V2 tasks with isolated Chrome runs, request interception, five-layer traces, and agentic judging.

馃弳 EchoZ-1.0: Elo 1035.7 路 Prediction Intelligence

Echo Leaderboard

A dynamic evaluation engine for AI prediction systems, featuring multi-point aligned Elo ranking, three-track data collection, and adaptive scheduling across diverse domains including finance, politics, crypto, sports, and esports.

馃弳 Qwen3.8-Max: 91.3% 路 Visual Reasoning

BabyVision

A benchmark for visual reasoning that challanges frontier MLLMs yet 3-year-olds can solve.