LLM Benchmark Table

Compare AI Model Performance Across Multiple Metrics

Model	TOTAL	Pass	Refine	Fail	Refusal	$ mToK	Reason	STEM	Utility	Code	Censor
Model Alpha	100	80	10	5	5	0.5	85	90	75	70	60
Model Beta	95	70	15	8	2	0.7	80	85	80	65	50
Model Gamma	110	90	8	7	5	0.4	88	92	78	75	55
Model Delta	88	65	12	10	1	0.9	75	80	70	60	45

Performance Overview Chart (Placeholder for Visual Data Representation)

Frequently Asked Questions

What does TOTAL represent?

TOTAL represents the overall score of the model across all tested metrics.

How is Pass rate calculated?

Pass rate is the percentage of tasks the model completed successfully without needing refinement.

What does $ mToK mean?

$ mToK refers to the cost in millions per thousand tokens processed by the model.