BenchLeader
Reka AIOpen weights: RekaAI/reka-flash-3Reasoning modelAuto-detected

Reka Flash 3

Best configuration ranks #541 of 610 on the BenchLeader Index at 38.9 ±3.5.

Blended price
$0.350/M
$0.200 in · $0.800 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 index39
  2. Agents & tools34
  3. Knowledge33
  4. Instruction following34
  5. Long context25
  6. Composite36

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 index52.9%#393Artificial Analysis
Humanity's Last Exam (AA)not in index4.4%#425Artificial Analysis

Agents & tools

Knowledge

BenchmarkdefaultSource
AA-Omniscience-65.1#423Artificial Analysis

Instruction following

BenchmarkdefaultSource
IFBench30.4%#343Artificial Analysis

Long context

BenchmarkdefaultSource
AA-LCR0.0%#419Artificial Analysis

Composite

BenchmarkdefaultSource
AA Intelligence Index5.7#479Artificial 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$0.0003
Summarise a 30-page report12,000 / 600$0.0029
Code edit6,000 / 1,500$0.0024
Agentic coding session60,000 / 4,000$0.015
Structured extraction2,000 / 200$0.0006

See also

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