Beta version: *Information might not be fully accurate. Please report any discrepancies.

IFBench

ifbenchAgents & ToolsHigher is better

Artificial Analysis IFBench. Evaluates precise instruction following with constraints.

Loading chart…

Loading leaderboard…

Benchmark Info

Current SOTA

Best Open Source

Median Score53.7%

DistributionBalanced

Score Distribution

P10

P50

P90

0100%

AgentBench

agentbench

IFEval

ifeval

Inverse IFEval

ifeval-inverse

Verified AdvancedIF

verified-advancedif

MultiChallenge

multichallenge

Terminal-Bench 2.0

terminal-bench