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 Sonnet 5Gemma 3 12B
Claude Sonnet 5

Anthropic

Gemma 3 12B

Google

Pricing
Input $/1M$2.00$0.05
Output $/1M$10.00$0.15
Blended $/1M (3:1)$4.00$0.075
Cache read $/1M$0.20
Cache write $/1M$2.50
Quality indices
Intelligence Index53.45.5
Coding Index71.55.8
Math Index18.3
Agentic Index46.70.3
Speed & latency
Output speed (tok/s)860
Time to first token93.73s0s
Time to first answer token93.73s0s
Benchmarks (Artificial Analysis)
MMLU-Pro59.5%
GPQA Diamond91.1%34.9%
Humanity's Last Exam39.6%4.8%
LiveCodeBench13.7%
SciCode53.6%17.4%
MATH-50085.3%
AIME22.0%
AIME 202518.3%
IFBench36.7%
AA-LCR (Long Context Reasoning)70.7%6.7%
Terminal-Bench Hard0.8%
Terminal-Bench 2.180.5%0.0%
τ²-Bench (Telecom)10.8%
τ³-Bench Banking28.2%0.8%
Design Arena (Elo)
Website1,314
Web apps1,305
Full-stack1,292
Android native1,255
UI components1,317
Data viz1,281
SVG1,239
3D1,321
Game dev1,345
Agentic game dev1,241
Godot game dev1,257
HTML slides1,233
Python→PPTX slides1,252
Code categories1,312
ASCII art1,230
Specs
Context window1M131K
Max output tokens128K16K
Input modalitiestext, image, filetext, image
TokenizerClaudeGemini