Compare AI models

Decision facts

• Qwen2.5 72B Instruct is #141; Qwen3.8 Max (0902) is #32 on the same current language board.
• Index scores are 20.5 for Qwen2.5 72B Instruct and 67.5 for Qwen3.8 Max (0902). 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.
• Evidence differs: Qwen2.5 72B Instruct is Measured; Qwen3.8 Max (0902) is Estimated.
• Listed output price per 1M tokens is $0.40 for Qwen2.5 72B Instruct and $6.00 for Qwen3.8 Max (0902). For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $0.92 for Qwen2.5 72B Instruct versus $7.00 for Qwen3.8 Max (0902). Qwen2.5 72B Instruct costs 86.9% 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 32,768 tokens for Qwen2.5 72B Instruct and 1,000,000 for Qwen3.8 Max (0902).
• Weight access differs: Qwen2.5 72B Instruct is restricted; Qwen3.8 Max (0902) is none.
• ModelCap currently lists 2 providers for Qwen2.5 72B Instruct and 1 for Qwen3.8 Max (0902).

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