BenchLeader
CohereOpen weightsReasoning modelAuto-detected

Command A Plus 05

Not yet ranked: too few independent results so far. Released 20 May 2026.

Blended price
$4.38/M
$2.50 in · $10.00 out
Output speed
First answer
Context
128k
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Coding41

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.

Coding

BenchmarkdefaultSource
SciCode37.9%#122SciCode

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$0.0040
Summarise a 30-page report12,000 / 600$0.036
Code edit6,000 / 1,500$0.030
Agentic coding session60,000 / 4,000$0.190
Structured extraction2,000 / 200$0.0070

See also

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