Other · model
Sarvam 105B benchmark scores
As of 2026-11-22, Sarvam 105B (Other) has measured scores on 16 benchmarks and estimated scores on 64 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 73.8% | measured |
| AA-HLE | 11.0% | measured |
| AA-SciCode | 30.5% | estimated ± 3.8 pp, high confidence |
| AIME26 | 91.0% | estimated ± 1.0 pp, low confidence |
| Artificial Analysis Intelligence Index | 12.2% | estimated ± 3.2 pp, medium confidence |
| CritPt | 0.0% | measured |
| GDPval-AA | 19.4% | estimated ± 7.8 pp, medium confidence |
| GPQA Diamond | 78.7% | measured |
| HLE | 13.7% | estimated ± 6.8 pp, medium confidence |
| IFBench | 53.8% | estimated ± 6.6 pp, low confidence |
| AA-LCR | 0.0% | measured |
| AA-Omniscience Accuracy | 28.5% | estimated ± 2.4 pp, high confidence |
| SciCode | 29.9% | estimated ± 3.6 pp, high confidence |
| BrowseComp | 49.5% | measured |
| HLE w/ tools | 36.4% | estimated ± 5.6 pp, medium confidence |
| IFEval | 84.8% | measured |
| LongBench v2 | 51.2% | estimated ± 5.5 pp, low confidence |
| VITA-Bench | 18.6% | estimated ± 6.7 pp, low confidence |
| τ²-bench results | 46.8% | measured |
| AA Coding Index | 23.9% | estimated ± 5.3 pp, medium confidence |
| APEX-Agents-AA | 8.1% | estimated ± 4.4 pp, medium confidence |
| BioMysteryBench (human-difficult) | 0.0% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 75.2% | estimated ± 6.8 pp, low confidence |
| SWE-bench Verified | 45.0% | measured |
| BFCL v4 | 41.2% | estimated ± 8.5 pp, low confidence |
| CyberGym | 35.4% | estimated ± 7.3 pp, low confidence |
| JobBench | 2.1% | estimated ± 10.7 pp, low confidence |
| SWE-bench Pro | 19.7% | estimated ± 4.5 pp, medium confidence |
| WideResearch | 59.8% | estimated ± 3.5 pp, medium confidence |
| AA Agentic Index | 12.6% | estimated ± 8.2 pp, medium confidence |
| MMLU-Pro | 81.7% | measured |
| AA-IFBench | 34.4% | measured |
| AIME 2024 | 91.0% | estimated ± 2.6 pp, high confidence |
| BBH | 96.3% | estimated ± 2.4 pp, low confidence |
| GPQA | 78.6% | estimated ± 1.2 pp, high confidence |
| MMLU | 90.6% | measured |
| Vals GPQA Diamond | 78.4% | estimated ± 2.1 pp, medium confidence |
| Vals LiveCodeBench | 49.9% | estimated ± 10.0 pp, low confidence |
| Vals MMLU-Pro | 80.1% | estimated ± 1.6 pp, medium confidence |
| MATH-500 | 98.6% | measured |
| MMMLU | 83.3% | estimated ± 0.7 pp, medium confidence |
| AIME 2025 | 88.3% | measured |
| AA MMLU-Pro | 85.7% | estimated ± 1.3 pp, medium confidence |
| Gert Labs | 28.9% | estimated ± 11.1 pp, low confidence |
| AA Harvey LAB v1.0 | 82.7% | estimated ± 9.8 pp, low confidence |
| ARC-AGI-1 | 74.7% | estimated ± 7.8 pp, low confidence |
| ARC-AGI-2 | 30.5% | estimated ± 14.4 pp, low confidence |
| FrontierMath v2 (Tiers 1-3) | 26.7% | estimated ± 11.2 pp, low confidence |
| MazeBench | 0.0% | estimated ± 6.4 pp, low confidence |
| OSWorld-Verified | 51.2% | estimated ± 5.7 pp, low confidence |
| Vals SWE-bench | 29.1% | estimated ± 6.6 pp, low confidence |
| CWE-Bench-AA | 34.7% | estimated ± 6.4 pp, medium confidence |
| AA Cyber Index | 25.7% | estimated ± 13.5 pp, low confidence |
| AA ITBench | 31.0% | estimated ± 4.1 pp, medium confidence |
| HLE w/o tools | 12.7% | estimated ± 4.3 pp, high confidence |
| Toolathlon-Verified | 34.0% | estimated ± 7.4 pp, low confidence |
| ARC-AGI-3 | 1.3% | estimated ± 9.8 pp, low confidence |
| ARC-AGI-3 Provider Adapter | 2.7% | estimated ± 7.3 pp, low confidence |
| Vibe Code Bench | 0.0% | estimated ± 8.4 pp, low confidence |
| C-Eval | 90.8% | estimated ± 1.1 pp, low confidence |
| Claw-Eval | 62.7% | estimated ± 6.4 pp, low confidence |
| HMMT Feb 2025 | 85.8% | measured |
| HMMT Feb 2026 | 81.3% | estimated ± 1.2 pp, low confidence |
| HMMT Nov 2025 | 85.8% | measured |
| MCP Atlas | 48.8% | estimated ± 10.3 pp, low confidence |
| MMAnswerBench | 77.7% | estimated ± 2.0 pp, low confidence |
| MMLU-Redux | 91.4% | estimated ± 0.6 pp, high confidence |
| SuperGPQA | 63.9% | estimated ± 12.4 pp, low confidence |
| Toolathlon | 24.4% | estimated ± 4.2 pp, medium confidence |
| HealthBench Hard | 0.0% | estimated ± 8.6 pp, low confidence |
| MMLU-Pro (Arcee) | 82.7% | estimated ± 3.4 pp, medium confidence |
| React Native Evals | 63.9% | estimated ± 2.7 pp, low confidence |
| ResearchClawBench | 12.3% | estimated ± 2.3 pp, low confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| SimpleQA | 25.5% | estimated ± 7.8 pp, medium confidence |
| LiveCodeBench | 53.9% | estimated ± 14.6 pp, low confidence |
| MRCRv2 | 22.8% | estimated ± 11.3 pp, low confidence |
| BioBasicsBench | 33.9% | estimated ± 10.7 pp, low confidence |
| LifeVal | 16.3% | estimated ± 1.6 pp, low confidence |
| PinchBench | 87.6% | estimated ± 3.6 pp, medium confidence |