Codestral 25.08 vs Qwen3 Omni 30B A3B Thinking
Price and specs side by side · USD per 1M tokens · verified 2026-11-22
// side by side
| metric | Codestral 25.08 | Qwen3 Omni 30B A3B Thinking |
|---|---|---|
| input / 1M | $0.30 | $0.25 |
| output / 1M | $0.90 | $0.97 |
| blended (3:1) | $0.45 | $0.43 |
| cache read / 1M | $0.03 | — |
| context | 256K | 65.5K |
| max output | 8.2K | 16.4K |
| input types | text · pdf | text · image · video · audio |
| reasoning | no | yes |
| tool calling | yes | yes |
| structured output | yes | yes |
| open weights | no | yes |
| released | 2025-07-30 | 2025-09-22 |
| knowledge cutoff | n/a | n/a |
| lab | Mistral | Alibaba (Qwen) |
// monthly cost scenarios
| workload | tokens in / out | Codestral 25.08 | Qwen3 Omni 30B A3B Thinking |
|---|---|---|---|
| Chatbot | 10M / 2M | $4.80 | $4.44 |
| RAG app | 50M / 5M | $19.50 | $17.35 |
| Coding agent | 20M / 10M | $15.00 | $14.70 |
// verdict
- Qwen3 Omni 30B A3B Thinking is 4.4% cheaper than Codestral 25.08 on a blended 3:1 input/output basis.
- Codestral 25.08 has the larger context window (256K vs 65.5K tokens).
- Only Qwen3 Omni 30B A3B Thinking supports built-in reasoning.
- Only Qwen3 Omni 30B A3B Thinking accepts image input.
- Chatbot (10M input / 2M output tokens per month): Qwen3 Omni 30B A3B Thinking costs $4.44 vs $4.80 for Codestral 25.08, saving $0.36 per month.
- RAG app (50M input / 5M output tokens per month): Qwen3 Omni 30B A3B Thinking costs $17.35 vs $19.50 for Codestral 25.08, saving $2.15 per month.
- Coding agent (20M input / 10M output tokens per month): Qwen3 Omni 30B A3B Thinking costs $14.70 vs $15.00 for Codestral 25.08, saving $0.30 per month.
// Codestral 25.08 or Qwen3 Omni 30B A3B Thinking: which one to choose
- Choose Qwen3 Omni 30B A3B Thinking for cost-sensitive, high-volume workloads.
- Choose Codestral 25.08 for long documents and large codebases (256K-token context).
- Choose Qwen3 Omni 30B A3B Thinking if you need built-in reasoning.
- Choose Qwen3 Omni 30B A3B Thinking if you need image input.
- Choose Qwen3 Omni 30B A3B Thinking if you want open weights you can self-host.