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 4.5Gemini 3.7 Flash
Claude Haiku 4.5

Anthropic

Gemini 3.7 Flash

Google

Pricing
Input $/1M$1.00$0.75
Output $/1M$5.00$3.75
Blended $/1M (3:1)$2.00$1.50
Cache read $/1M$0.10$0.075
Cache write $/1M$1.25$0.042
Quality indices
Intelligence Index16.939.1
Coding Index43.976.1
Math Index
Agentic Index835.2
Speed & latency
Output speed (tok/s)418
Time to first token14.77s
Time to first answer token14.77s
Benchmarks (Artificial Analysis)
GPQA Diamond94.5%
Humanity's Last Exam47.9%
SciCode57.2%
AA-LCR (Long Context Reasoning)81.7%
Terminal-Bench 2.185.8%
τ³-Bench Banking32.8%
Design Arena (Elo)
Website1,1351,315
Web apps1,235
Full-stack1,222
Mobile apps1,258
Android native1,253
UI components1,1151,312
Data viz1,1391,326
SVG1,049
3D1,1011,337
Game dev1,1201,350
Agentic game dev1,235
Code categories1,1301,323
ASCII art1,1591,271
Specs
Context window200K1.05M
Max output tokens64K66K
Input modalitiestext, image, filetext, image, video, file, audio
TokenizerClaudeGemini