Falcon-H1R-7B — benchmark results

TII's 7B hybrid Transformer-Mamba2 reasoning model with a 256K context (January 2026). Provider: TII. Released 2026-01-06. Access: Open.

Unified ELO 1579 ± 24, rank #533 of 1776 rated models, from 44 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA LiveCodeBench72.38Pass@1 (%)83.2
AA AIME 202580Accuracy (%)75
AA IFBench54.42Accuracy (%)65.4
CritPt0.3Accuracy (self-reported)65
AA Humanity's Last Exam10.8Accuracy (%)64.6
MathArena - SMT 202585.85Accuracy (%)60.7
AA CritPt0.29Accuracy (%)58.7
MathArena - AIME 202586.67Accuracy (%)55.6
MathArena - HMMT Feb 202584.17Accuracy (%)55.4
AA GPQA Diamond66.06Accuracy (%)45.8
AA Omniscience - Science, Engineering & Mathematics24.3Accuracy (%)42.9
AA MMLU-Pro72.54Accuracy (%)41.7

Interactive version: theaggregate.ai/model?slug=falcon-h1r-7b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.