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 Fable 5GPT-5.6 Terra
Claude Fable 5

Anthropic

GPT-5.6 Terra

OpenAI

Pricing
Input $/1M$10.00$2.50
Output $/1M$50.00$15.00
Blended $/1M (3:1)$20.00$5.63
Cache read $/1M$1.00$0.25
Cache write $/1M$12.50$3.13
Quality indices
Intelligence Index59.955
Coding Index76.576.7
Math Index
Agentic Index52.847.4
Speed & latency
Output speed (tok/s)70149
Time to first token58.27s104.73s
Time to first answer token58.27s104.73s
Benchmarks (Artificial Analysis)
GPQA Diamond92.6%92.5%
Humanity's Last Exam53.3%41.8%
SciCode60.2%53.9%
IFBench63.5%71.2%
AA-LCR (Long Context Reasoning)70.0%74.0%
Terminal-Bench Hard62.9%57.6%
Terminal-Bench 2.184.6%88.0%
τ²-Bench (Telecom)98.5%86.3%
τ³-Bench Banking26.8%31.8%
Design Arena (Elo)
Website1,330
Web apps1,318
Full-stack1,315
Mobile apps1,265
Android native1,322
UI components1,372
Data viz1,358
SVG1,352
3D1,362
Game dev1,396
Agentic game dev1,308
Godot game dev1,304
HTML slides1,259
Python→PPTX slides1,305
Agentic HTML slides1,254
Agentic slides (HTML)1,252
Code categories1,341
ASCII art1,373
Specs
Context window1M1.05M
Max output tokens128K128K
Input modalitiestext, image, filefile, image, text
TokenizerClaudeGPT