Qwen 2.5 72 B Instruct vs Qwen 3.5 Plus: cost comparison
Prices updated 2026-11-06 · computed from live provider data
Short answer: Qwen 2.5 72 B Instruct is cheaper on a
3:1 input/output blend — $0.370 per million tokens versus
$0.823, a 55% difference.
Qwen 3.5 Plus offers the larger context window at 983,616 tokens, against 32,768 for Qwen 2.5 72 B Instruct.
Head to head
| Qwen 2.5 72 B Instruct | Qwen 3.5 Plus | |
|---|---|---|
| Input, per 1M tokens | $0.360 | $0.360 |
| Output, per 1M tokens | $0.400 | $2.21 |
| Blended 3:1, per 1M | $0.370 | $0.823 |
| Context window | 32,768 | 983,616 |
| Vendor | qwen | empiriolabs |
Monthly cost at real volumes
| Workload (input / output) | Qwen 2.5 72 B Instruct | Qwen 3.5 Plus |
|---|---|---|
| 1M in / 250K out | $0.460 | $0.912 |
| 10M in / 2.5M out | $4.60 | $9.12 |
| 100M in / 25M out | $46.00 | $91.25 |
Which should you pick?
Pick Qwen 3.5 Plus when you need the longer context window (983,616 vs 32,768 tokens). Pick Qwen 2.5 72 B Instruct when price per token dominates your workload — high-volume classification, extraction, or batch summarisation.