BREAKING
Claude Sonnet 5 wins efficiency test
Tokens used per task
Sonnet 515047
Opus 4.823063
Sonnet 4.625824
GPT-5.531152
0$
Sonnet 5
0$
Opus 4.8
0$
GPT-5.5
What the test asked models to build
1Car hits wall
2Wrecking ball
3Catapult
Strengths versus open questions
Sonnet 5 edgePro
Most agentic Sonnet
~33% cheaper than 4.6
Strong token efficiency
Skeptics sayCaution
One-off demo, not general
Weaker graphics detail
Failure cases untested
Efficiency claim needs more tests
AI NEWS BLITZ
Anthropic's Claude Sonnet 5 topped a physics code test using the fewest tokens.