Compare / gpt-oss-20b vs Magistral Small 1.2
gpt-oss-20b vs Magistral Small 1.2
Updated November 2026
Verdict
gpt-oss-20b scores higher on the Small LM Index: 64.0 vs 55.9. On the 3 benchmarks both labs report, gpt-oss-20b wins 2 and Magistral Small 1.2 wins 1. gpt-oss-20b is smaller: it needs about 13.8 GB of memory at 4-bit, against 16.9 GB for Magistral Small 1.2. Only gpt-oss-20b has a hosted API on OpenRouter ($0.020 per million tokens); the other has to be self-hosted.
Pick gpt-oss-20b if…
- you want the strongest model overall
- you need math and reasoning
- you have limited memory (about 13.8 GB at 4-bit)
- you want a hosted API rather than running it yourself
- you work with long documents
- you want the more widely used model (more quantized builds and community support)
Pick Magistral Small 1.2 if…
- you mostly write code
- you need image input
Specifications
| gpt-oss-20b | Magistral Small 1.2 | |
|---|---|---|
| Small LM Index | 64.0 (likely 60–67) | 55.9 (likely 51–60) |
| Lab | OpenAI · United States | Mistral AI · Europe |
| Released | 2025-08-05 | 2025-09-17 |
| Parameters | 20.9B (3.6B active, MoE) | 24B (dense) |
| Memory at 4-bit | ~13.8 GB | ~16.9 GB |
| Context window | 131K tokens | 128K tokens |
| Input | text | text, image |
| Reasoning | Always thinks | Always thinks |
| License | Apache-2.0 | Apache-2.0 |
| API price / 1M tokens | $0.020 ($0.011 in, $0.045 out) | Self-host only |
| Hugging Face downloads (30 days) | 6M | 29.7K |
Benchmarks
Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.
| Benchmark | gpt-oss-20b | Magistral Small 1.2 |
|---|---|---|
| MMLU-Pro Knowledge | 75.0* | ≈ 76 |
| GPQA Diamond Knowledge | 71.5 | 70.1 |
| Humanity's Last Exam Reasoning | 10.9 | ≈ 9 |
| AIME Math | 91.7 | 77.3 |
| LiveCodeBench Coding | 61.0* | 70.9 |
| SWE-bench Verified Coding | 60.7 | ≈ 39 |
| IFBench Instruction following | 65.0* | — |
Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.
Questions
Which is better, gpt-oss-20b or Magistral Small 1.2?
gpt-oss-20b scores higher on the Small LM Index: 64.0 vs 55.9. On the 3 benchmarks both labs report, gpt-oss-20b wins 2 and Magistral Small 1.2 wins 1.
Which is cheaper to run?
Only gpt-oss-20b has a hosted API on OpenRouter ($0.020 per million tokens); the other has to be self-hosted. gpt-oss-20b is smaller: it needs about 13.8 GB of memory at 4-bit, against 16.9 GB for Magistral Small 1.2.
Which has the longer context window?
gpt-oss-20b has the longer context window: 131K tokens against 128K.
Can I use gpt-oss-20b and Magistral Small 1.2 commercially?
Both are released under the Apache-2.0 license. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.