Qwen 3 Coder 30 B A3 B Instruct vs Phi 4: cost comparison
Prices updated 2026-11-21 · computed from live provider data
Short answer: Phi 4 is cheaper on a
3:1 input/output blend — $0.087 per million tokens versus
$0.122, a 29% difference.
Qwen 3 Coder 30 B A3 B Instruct offers the larger context window at 262,144 tokens, against 16,384 for Phi 4.
Head to head
| Qwen 3 Coder 30 B A3 B Instruct | Phi 4 | |
|---|---|---|
| Input, per 1M tokens | $0.070 | $0.070 |
| Output, per 1M tokens | $0.280 | $0.140 |
| Blended 3:1, per 1M | $0.122 | $0.087 |
| Context window | 262,144 | 16,384 |
| Vendor | qwen | microsoft |
Monthly cost at real volumes
| Workload (input / output) | Qwen 3 Coder 30 B A3 B Instruct | Phi 4 |
|---|---|---|
| 1M in / 250K out | $0.140 | $0.105 |
| 10M in / 2.5M out | $1.40 | $1.05 |
| 100M in / 25M out | $14.00 | $10.50 |
Which should you pick?
Pick Qwen 3 Coder 30 B A3 B Instruct when you need the longer context window (262,144 vs 16,384 tokens). Pick Phi 4 when price per token dominates your workload — high-volume classification, extraction, or batch summarisation.