Mistral Code Fim Latest vs Qwen 3.6 27 B Fp 8: cost comparison
Prices updated 2026-10-29 · computed from live provider data
Short answer: Mistral Code Fim Latest is cheaper on a
3:1 input/output blend — $0.450 per million tokens versus
$1.04, a 57% difference.
Qwen 3.6 27 B Fp 8 offers the larger context window at 262,144 tokens, against 128,000 for Mistral Code Fim Latest.
Head to head
| Mistral Code Fim Latest | Qwen 3.6 27 B Fp 8 | |
|---|---|---|
| Input, per 1M tokens | $0.300 | $0.320 |
| Output, per 1M tokens | $0.900 | $3.20 |
| Blended 3:1, per 1M | $0.450 | $1.04 |
| Context window | 128,000 | 262,144 |
| Vendor | mistral | tensormesh |
Monthly cost at real volumes
| Workload (input / output) | Mistral Code Fim Latest | Qwen 3.6 27 B Fp 8 |
|---|---|---|
| 1M in / 250K out | $0.525 | $1.12 |
| 10M in / 2.5M out | $5.25 | $11.20 |
| 100M in / 25M out | $52.50 | $112 |
Which should you pick?
Pick Qwen 3.6 27 B Fp 8 when you need the longer context window (262,144 vs 128,000 tokens). Pick Mistral Code Fim Latest when price per token dominates your workload — high-volume classification, extraction, or batch summarisation.