Host Reka cut Qwen3.6 35B A3B input pricing by 33%
Published 26 September 2026
Price move, per 1M tokens
- Reka · input↓ 33%$0.15 → $0.10
- Reka · outputunchanged$1.00 → $1.00
- Reka · cache readunchanged$0.050 → $0.050
| host | rate | was | now | change |
|---|---|---|---|---|
| Reka | input | $0.15 | $0.10 | ↓ 33% |
| Reka | output | $1.00 | $1.00 | — |
| Reka | cache read | $0.050 | $0.050 | — |
input: ↓ 33%
If you buy
Reka's input rate for this model has dropped, so prompt-heavy workloads now cost less to run there; re-check your output-token spend before committing a larger budget.
The model
Qwen3.6 35B A3Bopen the model →
Open weights36B parameters262K context12 hosts
No leaderboard with a large enough field rates this model yet.
Qwen 3.6 is a multimodal model with a mixture-of-experts design that keeps only 3 billion parameters active for each word while drawing on 36 billion total. It carries a permissive Apache licence and is widely available from thirteen hosts, though no benchmark scores have been published yet.
Source
Read on openrouter.ai · we published this on 26 September 2026.
read the original at openrouter.ai ↗