IFBench Benchmark Results & Scores | Benchmark Registry

Versions of IFBench
Version	Released	Models	Results
IFBench OriginalLatest	July 3, 2025	17	18

Latest version: IFBench Original

Recently reported results for IFBench Original
Model	Score	Source
MAI-Thinking-1	69.0%	microsoft.ai for MAI-Thinking-1 on IFBench Original (opens in a new tab)AccuracyReported August 12, 2026Technical reportMicrosoft AISelf-reportedMicrosoftRecord for MAI-Thinking-1 on IFBench Original
Muse Glimmer 30B (high)	77.0%	dev.meta.ai for Muse Glimmer 30B (high) on IFBench Original (opens in a new tab)AccuracyReported August 10, 2026Model pageMetaSelf-reportedRecord for Muse Glimmer 30B on IFBench Original
Qwen3.8-Max	82.8%	huggingface.co for Qwen3.8-Max on IFBench Original (opens in a new tab)AccuracyReported August 2, 2026Model cardQwenSelf-reportedQwen TeamRecord for Qwen3.8-Max on IFBench Original
Inkling-Small (effort=0.99)	82.2%	thinkingmachines.ai for Inkling-Small (effort=0.99) on IFBench Original (opens in a new tab)AccuracyReported July 30, 2026Launch postThinking Machines LabSelf-reportedRecord for Inkling-Small on IFBench Original
Gemma 4 31B Instruct (thinking mode)	76.0%	arxiv.org for Gemma 4 31B Instruct (thinking mode) on IFBench Original (opens in a new tab)AccuracyReported July 24, 2026Technical reportGoogle DeepMindSelf-reportedRecord for Gemma 4 31B Instruct on IFBench Original