benchmark
.
space
benchmarks
rankings
compare
voices
transcripts
papers
articles
IFBench leaderboard
IFBench
3 models tested · Updated 2026-07-15 · Verified sources only
Inkling-Small
leads at
83.4%
1
Inkling-Small
Thinking Machines Lab ·
Blog/ThinkingMachines
· 2026-07-15
Beats Inkling (79.8) on instruction following.
83.4%
2
Grok 4.20
xAI ·
Artificial Analysis
· 2026-03-09
First place on instruction-following benchmark. From Artificial Analysis Intelligence Index v4.0.
83.0%
3
Inkling
Thinking Machines Lab ·
Blog/Thinking Machines Lab
· 2026-07-20
Instruction-following; slightly below Grok 4.20 (83.0) but above GPT-5.6 Sol (72.7).
79.8%