DeepSeek V4 Pro repriced across 2 hosts: Ionstream input down 22%, output up 25%
Published 24 September 2026
Price move, per 1M tokens
- Ionstream · input↓ 22%$0.50 → $0.39
- Ionstream · output↑ 25%$2.30 → $2.88
- Ionstream · cache read↑ 151%$0.035 → $0.088
- Wafer · input↓ 2%$0.395 → $0.387
- Wafer · outputunchanged$2.90 → $2.90
- Wafer · cache readunchanged$0.25 → $0.25
| host | rate | was | now | change |
|---|---|---|---|---|
| Ionstream | input | $0.50 | $0.39 | ↓ 22% |
| Ionstream | output | $2.30 | $2.88 | ↑ 25% |
| Ionstream | cache read | $0.035 | $0.088 | ↑ 151% |
| Wafer | input | $0.395 | $0.387 | ↓ 2% |
| Wafer | output | $2.90 | $2.90 | — |
| Wafer | cache read | $0.25 | $0.25 | — |
Ionstream input: ↓ 22%Ionstream output: ↑ 25%Wafer input: ↓ 2%
If you buy
On Ionstream, cached context now costs more to reuse, so check your cache-read share before the next top-up and lean on fresh input if that share is high.
The model
DeepSeek V4 Proopen the model →
Open weights1598.8B parameters1049K context28 hosts
intelligence36th of 168 writing32nd of 168 coding43rd of 168 agents31st of 55
DeepSeek V4 Pro is a large downloadable text model with a one-million-token request limit and strong mathematics scores. Its mixture-of-experts design keeps only 37 billion parameters active per token, making it more efficient to run than its total size suggests.
Source
Read on openrouter.ai · we published this on 24 September 2026.
read the original at openrouter.ai ↗