Skip to content
ModelCap

Model decision surface

Compare AI models

Start with GLM 4.5 Air and GLM 4.6V, 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

GLM 4.5 Air vs GLM 4.6V

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

Open 1200×630 evidence receipt
Factual comparison of GLM 4.5 Air and GLM 4.6V
Field
ModelCap position#112#109
Index score48.148.5
EvidenceMeasured2 public benchmark observations across 2 boardsMeasured2 public benchmark observations across 2 boards
Input / 1M$0.13$0.30
Output / 1M$0.85$0.90
Pricing statusfreshfresh
Context131K131K
Providers32
Weight accessOpen weightsOpen weights

Decision facts

  • GLM 4.5 Air is #112; GLM 4.6V is #109 on the same current language board.
  • Index scores are 48.1 for GLM 4.5 Air and 48.5 for GLM 4.6V. 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 $0.85 for GLM 4.5 Air and $0.90 for GLM 4.6V. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $0.69 for GLM 4.5 Air versus $1.05 for GLM 4.6V. GLM 4.5 Air costs 34.8% 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 131,072 tokens for GLM 4.5 Air and 131,072 for GLM 4.6V.
  • ModelCap currently lists 3 providers for GLM 4.5 Air and 2 for GLM 4.6V.

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.