Qwen 3.6 27 B vs Qwen 3 235 B A22 B Thinking 2507: cost comparison
Prices updated 2026-10-29 · computed from live provider data
Short answer: Qwen 3.6 27 B is cheaper on a
3:1 input/output blend — $1.01 per million tokens versus
$1.21, a 16% difference.
Qwen 3.6 27 B offers the larger context window at 262,144 tokens, against 128,000 for Qwen 3 235 B A22 B Thinking 2507.
Head to head
| Qwen 3.6 27 B | Qwen 3 235 B A22 B Thinking 2507 | |
|---|---|---|
| Input, per 1M tokens | $0.450 | $0.450 |
| Output, per 1M tokens | $2.70 | $3.50 |
| Blended 3:1, per 1M | $1.01 | $1.21 |
| Context window | 262,144 | 128,000 |
| Vendor | qwen | qwen |
Monthly cost at real volumes
| Workload (input / output) | Qwen 3.6 27 B | Qwen 3 235 B A22 B Thinking 2507 |
|---|---|---|
| 1M in / 250K out | $1.12 | $1.32 |
| 10M in / 2.5M out | $11.25 | $13.25 |
| 100M in / 25M out | $112 | $132 |
Which should you pick?
Pick Qwen 3.6 27 B when you need the longer context window (262,144 vs 128,000 tokens). Pick Qwen 3.6 27 B when price per token dominates your workload — high-volume classification, extraction, or batch summarisation.