Back

Cohere

Command A+

ReasoningOpen weightsModels.devHugging Face
Intelligence
13.1
Rank 342 of 744 measured models
Coding
27.8
Rank 189 of 298 measured models
Output speed
182 t/s
Rank 45 of 403 measured models
Context window
192K
Rank 674 of 1199 measured models
Cost per task
$0
Rank 1 of 193 measured models

Command A+ scores 13.1 on the Artificial Analysis Intelligence Index, ranking 342nd of 744 measured models. The closest models on this index are EXAONE 4.5 33B (13.2), DeepSeek R1 0528 (May '25) (13.1), and Qwen3.5 4B (13.1). With an output speed of 182 t/s, it is faster than DeepSeek R1 0528 (May '25) (33.9 t/s) and Qwen3.5 4B (19.2 t/s).

Artificial Analysis Indices
Default
Intelligence13.1
Rank 342 of 744 measured models
Coding27.8
Rank 189 of 298 measured models
Agentic1.0
Rank 158 of 189 measured models
Reasoning levels
Each point is a reasoning level of this model: more reasoning often raises the score but costs more. Select a point to open that level.
Intelligence Indexmost efficient ↗Output speed
Capability indexes
Performance in specific industries, measured by Artificial Analysis
Finance & Accounting10.6
Rank 181 of 218 measured models
Strategy & Ops4.1
Rank 206 of 218 measured models
Legal15.3
Rank 175 of 218 measured models
Healthcare & Medical11.7
Rank 76 of 82 measured models
Engineering18.7
Rank 165 of 217 measured models
Economics23.8
Rank 171 of 225 measured models
Standard Benchmarks
GPQA Diamond76.1%
Rank 256 of 619 measured models
HLE12.0%
Rank 293 of 664 measured models
SciCode38.5%
Rank 185 of 231 measured models
IFBench73.9%
Rank 38 of 450 measured models
AA-LCR52.7%
Rank 319 of 575 measured models
TerminalBench Hard25.0%
Rank 145 of 432 measured models
Terminal-Bench 2.122.8%
Rank 176 of 244 measured models
Terminal-Bench v40.5%
Rank 136 of 218 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy