Plan event · 21 Sept 2026
All plan events · 31 providers tracked
QwenCloudprice
What moved
Qwen3.8-Max proxied from qwen3.7-max-preview: $2.5 input / $7.5 output / $0.5 implicit cache per 1M → Qwen3.8-Max at the provider’s own published row: $2 input / $6 output / $0.25 implicit cache per 1M (explicit-cache rows not modeled) — affects Lite, Essential, Standard and Pro, which inherit the per-model map
Site data correction, not a claimed provider price cut: qwencloud.com/models/qwen3.8-max now publishes the model’s own rate row, so the calculator’s Qwen3.8-Max figure stops using the qwen3.7-max-preview proxy (which was cited as “3.8 not yet on the public PAYG table”) and roughly doubles Qwen3.8-Max Req/$1 on all four QwenCloud Token plans (Lite 37.5 → 71.75; the same correction applies to Essential, Standard and Pro). The provider’s internal per-token credit coefficients are console-only, so whether a subscriber’s credit consumption actually changed is not observable from the public pages — hence direction “neutral”.
Verified 2026-09-21 · source
Affected plans
- Lite Plan$8/mo
- Essential Plan$16/mo
- Standard Plan$25/mo
- Pro Plan$80/mo
More QwenCloud events
- 2026-09-21 · Launch — QwenCloud adds a new Essential tier to the Personal Token Plan at $16/mo with a limited-ti
- 2026-09-21 · Models — The docs "Supported models" allowlist for the Personal Token Plan grew from 7 to 10 text-c
- 2026-09-21 · Policy — The night-discount surfacing changed: the live docs list four night-discount models (up fr
- 2026-08-10 · Policy — QwenCloud cut the qwen3.8-max night discount from 80% extra to 50% off credits and tempora