gpt-6-luna
Benchmark profile · LiveBench 2.0 ·
Results for this model
- Overall benchmark score
- 66.7 / 100
- Cost per task
- $0.0801
- Cost per success
- $0.1200
- Coding
- 84.5 / 100
- Reasoning
- 81.8 / 100
- Browser use
- 70.3 / 100
Costs are based on the benchmark’s 471 shared tasks. Cost per success adjusts the reported cost by the model’s overall score.