Host Mancer 2 cut Qwen3.8 27B input pricing by 11%
Published 18 September 2026
Price move, per 1M tokens
- Mancer 2 · input↓ 11%$0.225 → $0.200
- Mancer 2 · outputunchanged$2.50 → $2.50
| host | rate | was | now | change |
|---|---|---|---|---|
| Mancer 2 | input | $0.225 | $0.200 | ↓ 11% |
| Mancer 2 | output | $2.50 | $2.50 | — |
input: ↓ 11%
If you buy
a lower input rate on this model at this host makes long-prompt work easier to justify, so re-check your cached-token and output rates before your next top-up.
The model
Qwen3.8 27Bopen the model →
Open weights27.8B parameters262K context20 hosts
intelligence63rd of 168 writing93rd of 168 coding45th of 168 agents33rd of 55
Qwen3.8 27B is a downloadable model you can run yourself, and its measured strength is maths and reasoning. Agentic recovery and tool use sit near the bottom of the field, so it suits a solver rather than an agent that has to get itself back on track.
Source
Read on openrouter.ai · we published this on 18 September 2026.
read the original at openrouter.ai ↗