Skip to content

Plan event · 23 Sept 2026

All plan events · 31 providers tracked

LLM Gateway DevPassprice

Price→NeutralProvider profile →

What moved

Cached read billed at the input rate on 6 rows; MiniMax M3, Nemotron 3 Ultra 550B and DeepSeek V4 Pro billed cached reads at 0 → Cached read billed at the gateway catalogue’s published cachedInputPrice on the mapping each row stores (0.20 / 0.20 / 0.01 / 0.50 / 0.0036 / 0.0028 restored; 0.06 / 0.10 / 0.0036 replace the zeros)

Calculator cache correction: the DevPass plan page publishes input/output only, but the gateway’s own catalogue publishes a cachedInputPrice per upstream mapping, so cached tokens bill at that published rate rather than at the input rate — and three rows that billed cache reads as free (a 100% discount) now carry the published rate.

Verified 2026-09-23 · source

Affected plans

More LLM Gateway DevPass events

Newsletter

Never miss a plan churn — get the weekly brief.

A plain weekly digest of every tracked price move, ceiling change and model-list update — pulled straight from the dated changelog ledger, with a one-line “worth switching?” note on each. Free. No spam, unsubscribe anytime.

Free · weekly · unsubscribe anytime