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 3 HaikuGPT-5.6 Terra
Claude 3 Haiku

Anthropic

GPT-5.6 Terra

OpenAI

Pricing
Input $/1M$0.25$2.50
Output $/1M$1.25$15.00
Blended $/1M (3:1)$0.50$5.63
Cache read $/1M$0.03$0.25
Cache write $/1M$0.30$3.13
Quality indices
Intelligence Index3.955
Coding Index76.7
Math Index
Agentic Index47.4
Speed & latency
Output speed (tok/s)0149
Time to first token0s104.73s
Time to first answer token0s104.73s
Benchmarks (Artificial Analysis)
GPQA Diamond37.4%92.5%
Humanity's Last Exam3.9%41.8%
LiveCodeBench15.4%
SciCode18.6%53.9%
MATH-50039.4%
AIME1.0%
IFBench36.1%71.2%
AA-LCR (Long Context Reasoning)21.0%74.0%
Terminal-Bench Hard0.8%57.6%
Terminal-Bench 2.188.0%
τ²-Bench (Telecom)21.1%86.3%
τ³-Bench Banking31.8%
Specs
Context window200K1.05M
Max output tokens4K128K
Input modalitiestext, imagefile, image, text
TokenizerClaudeGPT