Compare AI models

Decision facts

• Mercury 2.5 is #111; gpt-oss-120b is #114 on the same current language board.
• Index scores are 37.3 for Mercury 2.5 and 35.9 for gpt-oss-120b. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
• Evidence differs: Mercury 2.5 is Estimated; gpt-oss-120b is Measured.
• Listed output price per 1M tokens is $0.15 for Mercury 2.5 and $0.17 for gpt-oss-120b. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $0.15 for Mercury 2.5 versus $0.16 for gpt-oss-120b. Mercury 2.5 costs 2.5% less in this scenario. This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the selected endpoint before budgeting.
• Published context is 260,000 tokens for Mercury 2.5 and 131,072 for gpt-oss-120b.
• Weight access differs: Mercury 2.5 is none; gpt-oss-120b is open.
• ModelCap currently lists 1 providers for Mercury 2.5 and 18 for gpt-oss-120b.

These are separate published fields, not a synthetic winner. ModelCap does not collapse price, access, context, and capability evidence into a hidden recommendation score.