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 5Mistral Large
Claude Sonnet 5

Anthropic

Mistral Large

Mistral

Pricing
Input $/1M$2.00$2.00
Output $/1M$10.00$6.00
Blended $/1M (3:1)$4.00$3.00
Cache read $/1M$0.20$0.20
Cache write $/1M$2.50
Quality indices
Intelligence Index38.25.8
Coding Index71.5
Math Index
Agentic Index43.6
Speed & latency
Output speed (tok/s)820
Time to first token118.14s0s
Time to first answer token118.14s0s
Benchmarks (Artificial Analysis)
MMLU-Pro51.5%
GPQA Diamond91.1%35.1%
Humanity's Last Exam41.3%3.5%
LiveCodeBench17.8%
SciCode54.3%
MATH-50052.7%
AIME0.0%
AA-LCR (Long Context Reasoning)82.0%
Terminal-Bench 2.180.5%
τ³-Bench Banking37.3%
Design Arena (Elo)
Website1,288
Web apps1,260
Full-stack1,260
Mobile apps1,249
Android native1,240
UI components1,299
Data viz1,260
SVG1,212
3D1,289
Game dev1,313
Agentic game dev1,227
Godot game dev1,238
HTML slides1,220
Python→PPTX slides1,216
Code categories1,293
ASCII art1,226
Specs
Context window1M128K
Max output tokens128K102K
Input modalitiestext, image, filetext, file
TokenizerClaudeMistral