# Sarvam 105B Benchmark Scores & Performance

> Sarvam 105B by Sarvam scores 31.34/100 overall, ranking #162 out of 638 AI models.

Canonical page: https://benchlm.ai/models/sarvam-105b

Last updated: October 1, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Sarvam |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 128K |
| Overall Score | 31.34/100 |
| Overall Rank | #162 of 638 |

## Family & Coverage

- Family: Sarvam 105B
- Variant: base
- Benchmarks covered: 10 of 508
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [τ²-bench results](/benchmarks/tau2-bench) | 46.8% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |

## Reasoning Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AA-LCR](/benchmarks/lcr) | 0.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |
| [CritPt](/benchmarks/critpt) | 0.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |

## Knowledge Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 8.8% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 73.8% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |
| [AA-HLE](/benchmarks/aahle) | 11.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | -59.4% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 17.6% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 93.4% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |

## Instruction Following Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AA-IFBench](/benchmarks/aaifbench) | 34.4% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/sarvam-105b) |

## Other Sarvam Models

- [Sarvam 30B](/models/sarvam-30b) - Score: 29.11