What matters in AI.

Subscribe

Compute

The cost of a task on GLM-5.3 is $0.92 at Makora and $8.68 at Nebius

Makora charges $0.10 for 1M input tokens from cache, but Nebius charges $1.40 for all input, as Artificial Analysis found.

The same model costs about 9 times more per task on Nebius GLM-5.3 (max) costs $0.92 per task on Makora and $8.68 on Nebius. Cache hits and output price match, but cached input pricing differs: $0.10 against $1.40. ~9x COST PER TASK GLM-5.3 (max) Makora $0.92 Nebius $8.68 MAKORA NEBIUS Cache hit98%100% Output, per 1M$4.40$4.40 Cached input, per 1M$0.10$1.40 no discount
GLM-5.3 (max) costs $0.92 per task on Makora and $8.68 on Nebius. Cache hits and output price match, but input pricing differs.

Sources

  1. @ArtificialAnlys: Cache hit rates and Cost per Task by provider are now live on Artificial Analysis: Cost per Task for GLM-5.3 (max) varies ~9x across providers, driven by a combination of cache hit rx.com
AI MATTER · NEWS · AI MATTER · NEWS ·13 OCT2026

Posted

More in Compute

All Compute news