command-a-03-2025 vs sonar

Pricing, Performance & Features Comparison

Price unit:

command-a-03-2025

Authorcohere

Context Length256K

Reasoning

Providers1

ReleasedMar 2025

Knowledge Cutoff-

License-

Command A is our most performant model to date, excelling at tool use, agents, retrieval augmented generation (RAG), and multilingual use cases. Command A has a context length of 256K, only requires two GPUs to run, and has 150% higher throughput compared to Command R+ 08-2024.

Input$2.5

Output$10

Latency (p50)781ms

Output Limit-

Function Calling

JSON Mode

Input-

Output-

cohere

in$2.5out$10--

Latency (24h)

Success Rate (24h)

sonar

Authorperplexity

Context Length128K

Reasoning

Providers1

ReleasedFeb 2025

Knowledge CutoffApr 2024

License-

A lightweight, cost-effective search model optimized for quick, grounded answers with real-time web search.

Input$1

Output$1

Latency (p50)1.8s

Output Limit-

Function Calling

JSON Mode

Input-

Output-

perplexity

in$1out$1--