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 4.5Gemma 3 27B
Claude Haiku 4.5

Anthropic

Gemma 3 27B

Google

Pricing
Input $/1M$1.00$0.10
Output $/1M$5.00$0.30
Blended $/1M (3:1)$2.00$0.15
Cache read $/1M$0.10
Cache write $/1M$1.25
Quality indices
Intelligence Index29.67.4
Coding Index43.910.1
Math Index20.7
Agentic Index16.40.3
Speed & latency
Output speed (tok/s)0
Time to first token0s
Time to first answer token0s
Benchmarks (Artificial Analysis)
MMLU-Pro66.9%
GPQA Diamond42.8%
Humanity's Last Exam4.7%
LiveCodeBench13.7%
SciCode21.2%
MATH-50088.3%
AIME25.3%
AIME 202520.7%
IFBench31.8%
AA-LCR (Long Context Reasoning)5.7%
Terminal-Bench Hard3.8%
Terminal-Bench 2.14.5%
τ²-Bench (Telecom)10.5%
τ³-Bench Banking0.8%
Design Arena (Elo)
Website1,149
UI components1,142
Data viz1,160
SVG1,072
3D1,131
Game dev1,155
Code categories1,148
ASCII art1,182
Specs
Context window200K262K
Max output tokens64K
Input modalitiestext, image, filetext, image
TokenizerClaudeGemini