benchgap
Other · model

Sarvam 30B benchmark scores

As of 2026-11-22, Sarvam 30B (Other) has measured scores on 14 benchmarks and estimated scores on 61 more.

BenchmarkScoreSource
AA-GPQA Diamond63.3%measured
AA-HLE7.5%measured
AA-SciCode23.9%estimated ± 3.8 pp, medium confidence
AIME2686.2%estimated ± 1.6 pp, medium confidence
Artificial Analysis Intelligence Index9.7%estimated ± 3.2 pp, medium confidence
CritPt0.3%measured
GDPval-AA18.2%estimated ± 7.8 pp, low confidence
GPQA Diamond66.5%measured
HLE7.5%estimated ± 6.8 pp, medium confidence
IFBench53.6%estimated ± 6.6 pp, low confidence
AA-LCR0.0%measured
AA-Omniscience Accuracy16.4%estimated ± 2.4 pp, high confidence
SciCode23.4%estimated ± 3.6 pp, medium confidence
BrowseComp35.5%measured
HLE w/ tools34.7%estimated ± 5.6 pp, low confidence
IFEval80.3%estimated ± 2.8 pp, high confidence
LongBench v251.2%estimated ± 5.5 pp, low confidence
VITA-Bench6.1%estimated ± 6.7 pp, low confidence
τ²-bench results34.5%measured
AA Coding Index23.4%estimated ± 5.3 pp, medium confidence
APEX-Agents-AA6.9%estimated ± 4.4 pp, medium confidence
BioMysteryBench (human-difficult)6.5%estimated ± 6.8 pp, low confidence
DeepSearchQA74.5%estimated ± 6.8 pp, low confidence
SWE-bench Verified34.0%measured
BFCL v432.1%estimated ± 8.5 pp, low confidence
CyberGym23.2%estimated ± 7.3 pp, low confidence
JobBench0.0%estimated ± 10.7 pp, low confidence
SWE-bench Pro7.5%estimated ± 4.5 pp, medium confidence
WideResearch48.2%estimated ± 3.5 pp, medium confidence
AA Agentic Index7.6%estimated ± 8.2 pp, low confidence
MMLU-Pro80.0%measured
AA-IFBench26.5%measured
AIME 202477.6%estimated ± 7.1 pp, medium confidence
BBH90.0%estimated ± 2.4 pp, low confidence
GPQA66.3%estimated ± 1.2 pp, high confidence
MMLU85.1%measured
Vals GPQA Diamond62.4%estimated ± 3.2 pp, high confidence
Vals LiveCodeBench47.7%estimated ± 10.0 pp, low confidence
Vals MMLU-Pro72.0%estimated ± 2.4 pp, high confidence
MATH-50097.0%measured
MMMLU85.4%estimated ± 0.7 pp, medium confidence
AIME 202582.3%estimated ± 2.0 pp, high confidence
AA MMLU-Pro85.7%estimated ± 1.3 pp, medium confidence
Gert Labs28.8%estimated ± 11.1 pp, low confidence
AA Harvey LAB v1.078.9%estimated ± 9.8 pp, low confidence
ARC-AGI-175.0%estimated ± 7.8 pp, low confidence
ARC-AGI-231.1%estimated ± 14.4 pp, low confidence
MazeBench0.0%estimated ± 6.4 pp, low confidence
OSWorld-Verified49.6%estimated ± 5.7 pp, low confidence
Vals SWE-bench14.3%estimated ± 6.6 pp, low confidence
CWE-Bench-AA26.2%estimated ± 6.4 pp, low confidence
AA Cyber Index19.1%estimated ± 13.5 pp, low confidence
AA ITBench23.3%estimated ± 4.1 pp, medium confidence
HLE w/o tools8.9%estimated ± 4.3 pp, medium confidence
Toolathlon-Verified19.9%estimated ± 7.4 pp, low confidence
ARC-AGI-31.3%estimated ± 9.8 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
C-Eval91.1%estimated ± 1.1 pp, medium confidence
Claw-Eval62.7%estimated ± 6.4 pp, low confidence
HMMT Feb 202573.3%measured
HMMT Feb 202662.3%estimated ± 3.4 pp, medium confidence
HMMT Nov 202574.2%measured
MCP Atlas48.8%estimated ± 10.3 pp, low confidence
MMLU-Redux90.4%estimated ± 0.6 pp, high confidence
SuperGPQA61.6%estimated ± 12.4 pp, low confidence
Toolathlon16.0%estimated ± 4.2 pp, medium confidence
HealthBench Hard3.5%estimated ± 8.6 pp, low confidence
MMLU-Pro (Arcee)77.3%estimated ± 3.4 pp, medium confidence
React Native Evals58.9%estimated ± 2.7 pp, low confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
SimpleQA20.2%estimated ± 7.8 pp, medium confidence
LiveCodeBench43.6%estimated ± 14.6 pp, low confidence
MRCRv222.8%estimated ± 11.3 pp, low confidence
BioBasicsBench33.9%estimated ± 10.7 pp, low confidence
PinchBench85.1%estimated ± 3.6 pp, low confidence