Skip to content
ModelCap

Model decision surface

Compare AI models

Start with Qwen2.5 72B Instruct and Qwen3 30B A3B, or choose any two current ranked language models. Compare capability evidence, price, context, provider availability, and weight access without pretending one field decides every use case.

Current public data

Qwen2.5 72B Instruct vs Qwen3 30B A3B

Live dataset updated 9/22/2026, 10:37:09 PM UTC

Open 1200×630 evidence receipt
Factual comparison of Qwen2.5 72B Instruct and Qwen3 30B A3B
Field
ModelCap position#183#164
Index score26.033.6
EvidenceMeasured2 public benchmark observations across 2 boardsMeasured2 public benchmark observations across 2 boards
Input / 1M$0.36$0.12
Output / 1M$0.40$0.50
Pricing statusfreshfresh
Context33K131K
Providers22
Weight accessRestricted licenseOpen weights

Decision facts

  • Qwen2.5 72B Instruct is #183; Qwen3 30B A3B is #164 on the same current language board.
  • Index scores are 26.0 for Qwen2.5 72B Instruct and 33.6 for Qwen3 30B A3B. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
  • Both positions use Measured evidence.
  • Listed output price per 1M tokens is $0.40 for Qwen2.5 72B Instruct and $0.50 for Qwen3 30B A3B. 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 $0.49 for Qwen3 30B A3B. Qwen3 30B A3B costs 46.7% 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 131,072 for Qwen3 30B A3B.
  • Weight access differs: Qwen2.5 72B Instruct is restricted; Qwen3 30B A3B is open.

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

Each comparison page is a permanent, shareable URL with the same live figures as this tool: ModelCap Index position, API pricing, context window, provider count, weight access and every shared benchmark board.