Mistral Small 3.2 24 B Instruct vs GLM 5.3 Flash: cost comparison
Prices updated 2026-11-05 · computed from live provider data
Short answer: Mistral Small 3.2 24 B Instruct is cheaper on a
3:1 input/output blend — $0.106 per million tokens versus
$0.119, a 11% difference.
GLM 5.3 Flash offers the larger context window at 1,048,576 tokens, against 256,000 for Mistral Small 3.2 24 B Instruct.
Head to head
| Mistral Small 3.2 24 B Instruct | GLM 5.3 Flash | |
|---|---|---|
| Input, per 1M tokens | $0.075 | $0.075 |
| Output, per 1M tokens | $0.200 | $0.250 |
| Blended 3:1, per 1M | $0.106 | $0.119 |
| Context window | 256,000 | 1,048,576 |
| Vendor | mistralai | empiriolabs |
Monthly cost at real volumes
| Workload (input / output) | Mistral Small 3.2 24 B Instruct | GLM 5.3 Flash |
|---|---|---|
| 1M in / 250K out | $0.125 | $0.138 |
| 10M in / 2.5M out | $1.25 | $1.38 |
| 100M in / 25M out | $12.50 | $13.75 |
Which should you pick?
Pick GLM 5.3 Flash when you need the longer context window (1,048,576 vs 256,000 tokens). Pick Mistral Small 3.2 24 B Instruct when price per token dominates your workload — high-volume classification, extraction, or batch summarisation.