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 5gpt-oss-20b
Claude Sonnet 5

Anthropic

gpt-oss-20b

OpenAI

Pricing
Input $/1M$2.00$0.03
Output $/1M$10.00$0.13
Blended $/1M (3:1)$4.00$0.055
Cache read $/1M$0.20$0.03
Cache write $/1M$2.50
Quality indices
Intelligence Index53.414.9
Coding Index71.520.7
Math Index89.3
Agentic Index46.73.1
Speed & latency
Output speed (tok/s)86243
Time to first token93.73s0.39s
Time to first answer token93.73s8.61s
Benchmarks (Artificial Analysis)
MMLU-Pro74.8%
GPQA Diamond91.1%68.8%
Humanity's Last Exam39.6%9.8%
LiveCodeBench77.7%
SciCode53.6%34.4%
AIME 202589.3%
IFBench65.1%
AA-LCR (Long Context Reasoning)70.7%30.7%
Terminal-Bench Hard10.6%
Terminal-Bench 2.180.5%13.9%
τ²-Bench (Telecom)60.2%
τ³-Bench Banking28.2%7.0%
Design Arena (Elo)
Website1,314880
Web apps1,305
Full-stack1,292
Android native1,255
UI components1,317
Data viz1,281971
SVG1,239
3D1,321
Game dev1,345
Agentic game dev1,241
Godot game dev1,257
HTML slides1,233
Python→PPTX slides1,252
Code categories1,312
ASCII art1,230
Specs
Context window1M131K
Max output tokens128K131K
Input modalitiestext, image, filetext
TokenizerClaudeGPT