Tutorials, deep dives and product notes — built for developers.
Terminal-Bench 4.0: the tbench.ai public board remains led by Claude Opus 5.5 (64.8%). Adds Kimi K3.1 Pro (65.7%, Moonshot), Gemini 4 Carbon (64%, Artificial Analysis), GPT-6.1 Terra (50.3%) and MiniMax M3.1 Flash (38.1%) as separately labeled runs. Updated Oct 30, 2026.
FrontierBench v0.1 / Terminal-Bench 3.0: adds Snorkel AI's GLM-5.3 result (32.4% ±1.5, Claude Code) alongside Z.ai's distinct 28.3% launch result. Updated October 16, 2026.
MCP Atlas: adds Meta Muse Spark Open at 89.2% (xhigh, model-card evaluation), alongside Scale’s and other provider-specific results. Muse Spark 1.2 remains highest at 90.3%. No hosted API price is published for Spark Open. Updated Oct 23, 2026.