Compare AI models

Decision facts

• Qwen3.8 Max (0902) is #32; Qwen3 Coder Next is #103 on the same current language board.
• Index scores are 67.5 for Qwen3.8 Max (0902) and 40.2 for Qwen3 Coder Next. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
• Evidence differs: Qwen3.8 Max (0902) is Estimated; Qwen3 Coder Next is Measured.
• Listed output price per 1M tokens is $6.00 for Qwen3.8 Max (0902) and $0.80 for Qwen3 Coder Next. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $7.00 for Qwen3.8 Max (0902) versus $0.64 for Qwen3 Coder Next. Qwen3 Coder Next costs 90.9% 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 1,000,000 tokens for Qwen3.8 Max (0902) and 262,144 for Qwen3 Coder Next.
• Weight access differs: Qwen3.8 Max (0902) is none; Qwen3 Coder Next is open.
• ModelCap currently lists 1 providers for Qwen3.8 Max (0902) and 4 for Qwen3 Coder Next.

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