benchgap
Other · model

LLaDA2.2-mini benchmark scores

As of 2026-11-22, LLaDA2.2-mini (Other) has measured scores on 9 benchmarks and estimated scores on 30 more.

BenchmarkScoreSource
AA-GPQA Diamond55.3%estimated ± 4.4 pp, medium confidence
AA-HLE0.0%estimated ± 3.9 pp, medium confidence
AA-SciCode4.5%estimated ± 4.7 pp, medium confidence
AIME2635.1%measured
CritPt0.0%estimated ± 4.6 pp, medium confidence
GDPval-AA0.0%estimated ± 5.5 pp, low confidence
GPQA Diamond44.4%measured
HLE6.3%estimated ± 6.8 pp, low confidence
IFBench24.9%measured
AA-LCR33.6%estimated ± 12.5 pp, low confidence
LiveCodeBench v628.1%measured
AA-Omniscience Accuracy2.8%estimated ± 8.4 pp, low confidence
SciCode7.7%estimated ± 4.9 pp, medium confidence
BrowseComp55.5%estimated ± 13.0 pp, low confidence
IFEval77.7%estimated ± 3.6 pp, medium confidence
LongBench v235.0%measured
VITA-Bench35.0%estimated ± 10.7 pp, low confidence
τ²-bench results57.5%measured
AA Coding Index2.7%estimated ± 5.7 pp, low confidence
APEX-Agents-AA1.6%estimated ± 11.4 pp, low confidence
SWE-bench Verified57.5%estimated ± 9.7 pp, low confidence
BFCL v447.7%measured
JobBench23.7%estimated ± 9.3 pp, low confidence
AA Agentic Index11.5%estimated ± 12.7 pp, low confidence
MMLU-Pro55.5%estimated ± 5.4 pp, medium confidence
AA-IFBench56.7%estimated ± 7.4 pp, low confidence
GPQA44.1%estimated ± 1.2 pp, high confidence
MATH-50089.5%estimated ± 3.1 pp, low confidence
AIME 202539.2%estimated ± 1.8 pp, low confidence
OSWorld-Verified67.5%estimated ± 3.8 pp, medium confidence
Claw-Eval57.2%measured
HMMT Feb 202616.8%estimated ± 4.0 pp, medium confidence
MCP Atlas46.0%estimated ± 13.5 pp, low confidence
MCP-Tasks68.1%estimated ± 6.4 pp, low confidence
MMLU-Redux80.7%estimated ± 2.8 pp, high confidence
SuperGPQA23.7%estimated ± 13.9 pp, low confidence
MMLU-Pro (Arcee)63.9%estimated ± 3.4 pp, low confidence
ResearchClawBench15.7%estimated ± 1.3 pp, high confidence
PinchBench62.3%measured