Codestral 25.08 vs GPT OSS Safeguard 120B
Price and specs side by side · USD per 1M tokens · verified 2026-10-11
// side by side
| metric | Codestral 25.08 | GPT OSS Safeguard 120B |
|---|
| input / 1M | $0.30 | $0.15 |
| output / 1M | $0.90 | $0.60 |
| blended (3:1) | $0.45 | $0.2625 |
| cache read / 1M | $0.03 | — |
| context | 256K | 128K |
| max output | 8.2K | 16.4K |
| input types | text | text |
| reasoning | no | yes |
| tool calling | yes | yes |
| structured output | yes | yes |
| open weights | no | yes |
| released | 2025-07-30 | 2025-10-29 |
| knowledge cutoff | n/a | n/a |
| lab | Mistral | OpenAI |
// monthly cost scenarios
| workload | tokens in / out | Codestral 25.08 | GPT OSS Safeguard 120B |
|---|
| Chatbot | 10M / 2M | $4.80 | $2.70 |
| RAG app | 50M / 5M | $19.50 | $10.50 |
| Coding agent | 20M / 10M | $15.00 | $9.00 |
// verdict
- GPT OSS Safeguard 120B is 41.7% cheaper than Codestral 25.08 on a blended 3:1 input/output basis.
- Codestral 25.08 has the larger context window (256K vs 128K tokens).
- Only GPT OSS Safeguard 120B supports built-in reasoning.