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 4.5Kimi K2.6
Claude Haiku 4.5

Anthropic

Kimi K2.6

Moonshot AI

Pricing
Input $/1M$1.00$0.684
Output $/1M$5.00$3.42
Blended $/1M (3:1)$2.00$1.37
Cache read $/1M$0.10$0.144
Cache write $/1M$1.25
Quality indices
Intelligence Index29.644.2
Coding Index43.961.8
Math Index
Agentic Index16.430.3
Speed & latency
Output speed (tok/s)
Time to first token
Time to first answer token
Design Arena (Elo)
Website1,1491,304
Web apps1,268
Full-stack1,216
Mobile apps1,248
Android native1,294
UI components1,1421,305
Data viz1,1601,295
SVG1,0721,225
3D1,1311,333
Game dev1,1551,307
Agentic game dev1,157
Godot game dev1,168
HTML slides1,229
PPTX slides1,181
Python→PPTX slides1,180
Agentic slides1,187
Agentic HTML slides1,248
Agentic slides (HTML)1,252
Agentic slides (Python PPTX)1,186
Code categories1,1481,311
ASCII art1,1821,192
Specs
Context window200K262K
Max output tokens64K262K
Input modalitiestext, image, filetext, image
TokenizerClaudeOther