BenchLeader
SarvamOpen weightsReasoning modelAuto-detected

Sarvam 105B (high)

Best configuration ranks #509 of 610 on the BenchLeader Index at 40.4 ±4.7.

Blended price
$0.073/M
$0.040 in · $0.170 out
Output speed
First answer
Context
128k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index40
  2. Agents & tools35
  3. Knowledge36
  4. Instruction following38
  5. Long context25
  6. Composite40

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarkdefaultSource
GPQA Diamond (AA)not in index73.8%#254Artificial Analysis
Humanity's Last Exam (AA)not in index11.0%#238Artificial Analysis

Agents & tools

Knowledge

BenchmarkdefaultSource
AA-Omniscience-59.4#386Artificial Analysis

Instruction following

BenchmarkdefaultSource
IFBench34.4%#304Artificial Analysis

Long context

BenchmarkdefaultSource
AA-LCR0.0%#419Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index8.8#351Artificial Analysis

What a task costs

Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.

WorkloadTokens in / outCostWith cachingTime
Chat reply400 / 300$0.0001
Summarise a 30-page report12,000 / 600$0.0006
Code edit6,000 / 1,500$0.0005
Agentic coding session60,000 / 4,000$0.0031
Structured extraction2,000 / 200$0.0001

See also

Data as of 9 Sept 2026. Compare with another model.