DeepSeek V4 Flash repriced across 2 hosts: Inceptron input down 24%, Wafer output up 40%
Published 23 September 2026
Price move, per 1M tokens
- Wafer · input↓ 13%$0.100 → $0.087
- Wafer · output↑ 40%$0.25 → $0.35
- Wafer · cache read↑ 60%$0.050 → $0.080
- Inceptron · input↓ 24%$0.120 → $0.091
- Inceptron · output↑ 23%$0.500 → $0.613
- Inceptron · cache read↑ 20%$0.050 → $0.060
| host | rate | was | now | change |
|---|---|---|---|---|
| Wafer | input | $0.100 | $0.087 | ↓ 13% |
| Wafer | output | $0.25 | $0.35 | ↑ 40% |
| Wafer | cache read | $0.050 | $0.080 | ↑ 60% |
| Inceptron | input | $0.120 | $0.091 | ↓ 24% |
| Inceptron | output | $0.500 | $0.613 | ↑ 23% |
| Inceptron | cache read | $0.050 | $0.060 | ↑ 20% |
Wafer input: ↓ 13%Wafer output: ↑ 40%Inceptron input: ↓ 24%Inceptron output: ↑ 23%
If you buy
Output-heavy work on Wafer now costs more per token, so re-check your spend mix before your next top-up and weigh long generations against your budget.
The model
DeepSeek V4 Flashopen the model →
Open weights290.9B parameters1049K context33 hosts
intelligence65th of 168 writing57th of 168 coding64th of 168 agents20th of 55
DeepSeek V4 Flash is a downloadable text model with a permissive licence, built for long-document work where the bill matters more than peak quality. Its measured quality is weak on the LiveBench boards and mid-pack on the Arena boards, so treat it as a cheap workhorse rather than a quality leader.
Source
Read on openrouter.ai · we published this on 23 September 2026.
read the original at openrouter.ai ↗