GLM 5.3 Flash repriced across 4 hosts: OpenInference output down 38%, Sail Research output up 300%
Published 31 October 2026
Price move, per 1M tokens
- Sail Research · input↑ 233%$0.15 → $0.50
- Sail Research · output↑ 300%$0.50 → $2.00
- Sail Research · cache read↓ 33%$0.030 → $0.020
- Sail Research (US region) · input↑ 50%$0.150 → $0.225
- Sail Research (US region) · output↑ 50%$0.50 → $0.75
- Sail Research (US region) · cache read↑ 50%$0.030 → $0.045
- OpenInference · input↓ 17%$0.0207 → $0.0172
- OpenInference · output↓ 38%$0.0569 → $0.0355
- OpenInference · cache read↓ 9%$0.0172 → $0.0157
- Relace · input↓ 31%$0.0237 → $0.0163
- Relace · output↑ 30%$0.114 → $0.148
- Relace · cache read↓ 31%$0.0237 → $0.0163
- Wafer (US region) · input↓ 8%$0.048 → $0.044
- Wafer (US region) · outputunchanged$0.20 → $0.20
- Wafer (US region) · cache read↓ 9%$0.047 → $0.043
| host | rate | was | now | change |
|---|---|---|---|---|
| Sail Research | input | $0.15 | $0.50 | ↑ 233% |
| Sail Research | output | $0.50 | $2.00 | ↑ 300% |
| Sail Research | cache read | $0.030 | $0.020 | ↓ 33% |
| Sail Research (US region) | input | $0.150 | $0.225 | ↑ 50% |
| Sail Research (US region) | output | $0.50 | $0.75 | ↑ 50% |
| Sail Research (US region) | cache read | $0.030 | $0.045 | ↑ 50% |
| OpenInference | input | $0.0207 | $0.0172 | ↓ 17% |
| OpenInference | output | $0.0569 | $0.0355 | ↓ 38% |
| OpenInference | cache read | $0.0172 | $0.0157 | ↓ 9% |
| Relace | input | $0.0237 | $0.0163 | ↓ 31% |
| Relace | output | $0.114 | $0.148 | ↑ 30% |
| Relace | cache read | $0.0237 | $0.0163 | ↓ 31% |
| Wafer (US region) | input | $0.048 | $0.044 | ↓ 8% |
| Wafer (US region) | output | $0.20 | $0.20 | — |
| Wafer (US region) | cache read | $0.047 | $0.043 | ↓ 9% |
Sail Research output: ↑ 300%Sail Research cache read: ↓ 33%Sail Research (US region) all rates: ↑ 50%OpenInference output: ↓ 38%Relace input and cache read: ↓ 31%Relace output: ↑ 30%Wafer (US region) input: ↓ 8%
The model
GLM 5.3 Flashopen the model →
Open weights321.3B parameters1311K context43 hosts
intelligence24th of 185 writing44th of 185 coding20th of 185 agents42nd of 70
GLM 5.3 Flash is a downloadable model with a licence that allows commercial use, changes and redistribution, and a request capacity large enough that long documents need not be split up first. Its measured quality is uneven: strong on maths and coding, near the bottom of the field on instruction following.
Source
Read on openrouter.ai · we published this on 31 October 2026.
read the original at openrouter.ai ↗