Compare AI models

Decision facts

• Qwen2.5 72B Instruct is #250; Qwen3 235B A22B Instruct 2507 is #85 on the same current language board.
• Index scores are 27.5 for Qwen2.5 72B Instruct and 59.8 for Qwen3 235B A22B Instruct 2507. 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 $0.40 for Qwen2.5 72B Instruct and $0.80 for Qwen3 235B A22B Instruct 2507. 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 $0.68 for Qwen3 235B A22B Instruct 2507. Qwen3 235B A22B Instruct 2507 costs 26.1% 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 262,144 for Qwen3 235B A22B Instruct 2507.
• Weight access differs: Qwen2.5 72B Instruct is restricted; Qwen3 235B A22B Instruct 2507 is open.
• ModelCap currently lists 1 providers for Qwen2.5 72B Instruct and 4 for Qwen3 235B A22B Instruct 2507.

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