Qwen 3 235 B A22 B Fp 8 vs Qwen 3.5 397 B A17 B: cost comparison
Prices updated 2026-11-05 · computed from live provider data
Short answer: Qwen 3.5 397 B A17 B is cheaper on a
3:1 input/output blend — $0.387 per million tokens versus
$0.400, a 3% difference.
Qwen 3.5 397 B A17 B offers the larger context window at 258,048 tokens, against 32,768 for Qwen 3 235 B A22 B Fp 8.
Head to head
| Qwen 3 235 B A22 B Fp 8 | Qwen 3.5 397 B A17 B | |
|---|---|---|
| Input, per 1M tokens | $0.170 | $0.172 |
| Output, per 1M tokens | $1.09 | $1.03 |
| Blended 3:1, per 1M | $0.400 | $0.387 |
| Context window | 32,768 | 258,048 |
| Vendor | gmi | empiriolabs |
Monthly cost at real volumes
| Workload (input / output) | Qwen 3 235 B A22 B Fp 8 | Qwen 3.5 397 B A17 B |
|---|---|---|
| 1M in / 250K out | $0.443 | $0.430 |
| 10M in / 2.5M out | $4.42 | $4.30 |
| 100M in / 25M out | $44.25 | $43.00 |
Which should you pick?
Pick Qwen 3.5 397 B A17 B when you need the longer context window (258,048 vs 32,768 tokens). Pick Qwen 3.5 397 B A17 B when price per token dominates your workload — high-volume classification, extraction, or batch summarisation.