Codestral 25.08 vs Qwen3.8 Omni Flash
Price and specs side by side · USD per 1M tokens · verified 2026-10-11
// side by side
| metric | Codestral 25.08 | Qwen3.8 Omni Flash |
|---|---|---|
| input / 1M | $0.30 | $0.15 |
| output / 1M | $0.90 | $0.47 |
| blended (3:1) | $0.45 | $0.23 |
| cache read / 1M | $0.03 | $0.016 |
| context | 256K | 1M |
| max output | 8.2K | 131.1K |
| input types | text | text · image · audio · video |
| reasoning | no | yes |
| tool calling | yes | yes |
| structured output | yes | yes |
| open weights | no | no |
| released | 2025-07-30 | 2026-09-17 |
| knowledge cutoff | n/a | n/a |
| lab | Mistral | Alibaba (Qwen) |
// monthly cost scenarios
| workload | tokens in / out | Codestral 25.08 | Qwen3.8 Omni Flash |
|---|---|---|---|
| Chatbot | 10M / 2M | $4.80 | $2.44 |
| RAG app | 50M / 5M | $19.50 | $9.85 |
| Coding agent | 20M / 10M | $15.00 | $7.70 |
// verdict
- Qwen3.8 Omni Flash is 48.9% cheaper than Codestral 25.08 on a blended 3:1 input/output basis.
- Qwen3.8 Omni Flash has the larger context window (1M vs 256K tokens).
- Only Qwen3.8 Omni Flash supports built-in reasoning.
- Only Qwen3.8 Omni Flash accepts image input.