AiCostCompare

Compare models

Pick up to 4 models for a side-by-side spec sheet. Hover any dotted metric name for what the test measures. Best value in each row is highlighted.

Claude Haiku 4.5Command A
Claude Haiku 4.5

Anthropic

Command A

Cohere

Pricing
Input $/1M$1.00$2.50
Output $/1M$5.00$10.00
Blended $/1M (3:1)$2.00$4.38
Cache read $/1M$0.10
Cache write $/1M$1.25
Quality indices
Intelligence Index16.913.1
Coding Index43.927.8
Math Index
Agentic Index81
Speed & latency
Output speed (tok/s)220
Time to first token0.18s
Time to first answer token9.28s
Benchmarks (Artificial Analysis)
GPQA Diamond76.1%
Humanity's Last Exam12.0%
SciCode38.5%
IFBench73.9%
AA-LCR (Long Context Reasoning)52.7%
Terminal-Bench Hard25.0%
Terminal-Bench 2.122.8%
τ²-Bench (Telecom)80.7%
τ³-Bench Banking6.0%
Design Arena (Elo)
Website1,135
UI components1,115
Data viz1,139
SVG1,049
3D1,101
Game dev1,120
Code categories1,130
ASCII art1,159
Specs
Context window200K256K
Max output tokens64K8K
Input modalitiestext, image, filetext
TokenizerClaudeOther