Kimi K3 rose across 2 hosts, by up to 32% at Wafer (output)
Published 23 September 2026
Price move, per 1M tokens
- Wafer · input↑ 19%$2.10 → $2.49
- Wafer · output↑ 32%$10.95 → $14.50
- Wafer · cache read↑ 19%$0.21 → $0.25
- Relace · input↑ 15%$1.70 → $1.95
- Relace · output↑ 15%$8.50 → $9.75
- Relace · cache read↑ 15%$0.170 → $0.195
| host | rate | was | now | change |
|---|---|---|---|---|
| Wafer | input | $2.10 | $2.49 | ↑ 19% |
| Wafer | output | $10.95 | $14.50 | ↑ 32% |
| Wafer | cache read | $0.21 | $0.25 | ↑ 19% |
| Relace | input | $1.70 | $1.95 | ↑ 15% |
| Relace | output | $8.50 | $9.75 | ↑ 15% |
| Relace | cache read | $0.170 | $0.195 | ↑ 15% |
Wafer output: ↑ 32%Relace all rates: ↑ 15%
The model
Kimi K3open the model →
Open weights2779.9B parameters1049K context25 hosts
intelligence12th of 168 writing18th of 168 coding6th of 168 agents10th of 55
Kimi K3 is Moonshot AI's massive-scale mixture-of-experts model with 2.8 trillion total parameters and 104 billion active per word. It excels at reasoning and web-development tasks, and handles up to one million tokens in a single request, though its weights carry commercial restrictions.
Source
Read on openrouter.ai · we published this on 23 September 2026.
read the original at openrouter.ai ↗