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 4.5MiniMax M2
Claude Haiku 4.5

Anthropic

MiniMax M2

MiniMax

Pricing
Input $/1M$1.00$0.255
Output $/1M$5.00$1.02
Blended $/1M (3:1)$2.00$0.446
Cache read $/1M$0.10
Cache write $/1M$1.25
Quality indices
Intelligence Index16.918.6
Coding Index43.9
Math Index78.3
Agentic Index8
Speed & latency
Output speed (tok/s)0
Time to first token0s
Time to first answer token0s
Benchmarks (Artificial Analysis)
MMLU-Pro82.0%
GPQA Diamond77.7%
Humanity's Last Exam13.7%
LiveCodeBench82.6%
AIME 202578.3%
IFBench72.3%
AA-LCR (Long Context Reasoning)64.3%
Terminal-Bench Hard25.8%
τ²-Bench (Telecom)86.8%
Design Arena (Elo)
Website1,1351,155
UI components1,1151,152
Data viz1,1391,155
SVG1,0491,123
3D1,1011,126
Game dev1,1201,141
Code categories1,1301,150
ASCII art1,159
Specs
Context window200K205K
Max output tokens64K131K
Input modalitiestext, image, filetext
TokenizerClaudeOther