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 4.5Mistral Medium 3.5
Claude Haiku 4.5

Anthropic

Mistral Medium 3.5

Mistral

Pricing
Input $/1M$1.00$1.50
Output $/1M$5.00$7.50
Blended $/1M (3:1)$2.00$3.00
Cache read $/1M$0.10
Cache write $/1M$1.25
Quality indices
Intelligence Index29.629.9
Coding Index43.946.9
Math Index
Agentic Index16.419
Speed & latency
Output speed (tok/s)135
Time to first token0.80s
Time to first answer token15.59s
Benchmarks (Artificial Analysis)
GPQA Diamond74.8%
Humanity's Last Exam12.8%
SciCode39.6%
IFBench68.8%
AA-LCR (Long Context Reasoning)61.0%
Terminal-Bench Hard33.3%
Terminal-Bench 2.150.6%
τ²-Bench (Telecom)94.2%
τ³-Bench Banking14.4%
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, file
TokenizerClaudeMistral