Qwen3 235B A22B Instruct 2507
AlibabaA mid-sized Qwen3 mixture-of-experts model from Alibaba, priced well below the Coder variant.
Main uses: General text tasks at a very low cost per token, for teams already using Qwen3 Coder for code specifically.
What sets it apart: At $0.09/$0.55 per 1M tokens, under a fifth of Qwen3 Coder's output price for general (non-code) use.
In
$0.09
Out
$0.55
Blended
$0.21
Where to buy this model
4 access routes| Route | In | Out | Blended | vs |
|---|---|---|---|---|
| $0.09 | $0.35 | $0.15 | - | |
| $0.09 | $0.55 | $0.21 | +40% | |
| ● Venice fp8 | $0.15 | $0.75 | $0.30 | +100% |
| $0.14 | $0.80 | $0.31 | +107% |
● official API · ● cloud platform · ● inference host · USD per 1M tokens. Same model, different providers - price and quantization vary.
- Ctx
- 262K
- max output
- 16K
- Family
- qwen
- Modalities
- text
- Features
- json · tools
Live pricing, updated 2026-11-03.
Alibaba models at a similar price
| Model | In | Out | Ctx |
|---|---|---|---|
| Qwen3.5-9B | $0.17 | $0.25 | 262K |
| Qwen3 235B A22B Instruct 2507 · this model | $0.09 | $0.55 | 262K |
| Qwen3 30B A3B | $0.12 | $0.5 | 41K |
| Qwen3.8 Flash | $0.15 | $0.47 | 1M |
| Qwen3.8 Omni Flash | $0.15 | $0.47 | 1M |
| Qwen3 Coder 30B A3B Instruct | $0.15 | $0.6 | 262K |