GPT OSS 120B vs Ministral 3 3B
Price and specs side by side · USD per 1M tokens · verified 2026-11-21
// side by side
| metric | GPT OSS 120B | Ministral 3 3B |
|---|
| input / 1M | $0.037 | $0.10 |
| output / 1M | $0.17 | $0.10 |
| blended (3:1) | $0.0703 | $0.10 |
| cache read / 1M | $0.0185 | $0.01 |
| context | 131.1K | 131.1K |
| max output | 16.4K | 131.1K |
| input types | text | text · image · pdf |
| reasoning | yes | no |
| tool calling | yes | yes |
| structured output | yes | yes |
| open weights | yes | yes |
| released | 2025-08-05 | 2025-12-02 |
| knowledge cutoff | 2024-06 | n/a |
| lab | OpenAI | Mistral |
// monthly cost scenarios
| workload | tokens in / out | GPT OSS 120B | Ministral 3 3B |
|---|
| Chatbot | 10M / 2M | $0.71 | $1.20 |
| RAG app | 50M / 5M | $2.70 | $5.50 |
| Coding agent | 20M / 10M | $2.44 | $3.00 |
// verdict
- GPT OSS 120B is 29.7% cheaper than Ministral 3 3B on a blended 3:1 input/output basis.
- Only GPT OSS 120B supports built-in reasoning.
- Only Ministral 3 3B accepts image input.
- Chatbot (10M input / 2M output tokens per month): GPT OSS 120B costs $0.71 vs $1.20 for Ministral 3 3B, saving $0.49 per month.
- RAG app (50M input / 5M output tokens per month): GPT OSS 120B costs $2.70 vs $5.50 for Ministral 3 3B, saving $2.80 per month.
- Coding agent (20M input / 10M output tokens per month): GPT OSS 120B costs $2.44 vs $3.00 for Ministral 3 3B, saving $0.56 per month.
// GPT OSS 120B or Ministral 3 3B: which one to choose
- Choose GPT OSS 120B for cost-sensitive, high-volume workloads.
- Choose GPT OSS 120B if you need built-in reasoning.
- Choose Ministral 3 3B if you need image input.