Compare AI models

Decision facts

• GPT-6 Astra is #1; gpt-oss-120b is #110 on the same current language board.
• Index scores are 95.7 for GPT-6 Astra and 36.8 for gpt-oss-120b. Their published uncertainty intervals do not overlap, but the aggregate Index does not predict performance on every task; compare the relevant benchmark configurations before choosing.
• Both positions use Measured evidence.
• Listed output price per 1M tokens is $50.00 for GPT-6 Astra 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 $45.00 for GPT-6 Astra versus $0.16 for gpt-oss-120b. gpt-oss-120b costs 99.6% 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 1,050,000 tokens for GPT-6 Astra and 131,072 for gpt-oss-120b.
• Weight access differs: GPT-6 Astra is none; gpt-oss-120b is open.
• ModelCap currently lists 2 providers for GPT-6 Astra 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.