Model comparison
gpt-oss-120b vs Magistral 1.0 (Small & Medium)
- Input
- $0.039 / 1M tokens
- Output
- $0.18 / 1M tokens
- Context
- 131K
- Released
- 2025-08-05
- Input
- $0.50 / 1M tokens (Magistral Small); $2.00 / 1M tokens (Magistral Medium)
- Output
- $1.50 / 1M tokens (Magistral Small); $5.00 / 1M tokens (Magistral Medium)
- Context
- 128K
- Released
- 2025-06-10
Gpt-oss-120b scores higher on 3 of the 3 benchmarks both models report; gpt-oss-120b is cheaper on input tokens ($0 vs $0.5 per 1M, 12.8×); gpt-oss-120b has the larger context window (131K vs 128K).
Specs and pricing side by side
| Spec | gpt-oss-120b | Magistral 1.0 (Small & Medium) |
|---|---|---|
| Maker | OpenAI | Mistral AI |
| Released | 2025-08-05 | 2025-06-10 |
| Context window | 131K | 128K |
| Max output | 131K | Not separately published; recommended working… |
| Input price | $0.039 / 1M tokens | $0.50 / 1M tokens (Magistral Small); $2.00 / 1M tokens (Magistral Medium) |
| Output price | $0.18 / 1M tokens | $1.50 / 1M tokens (Magistral Small); $5.00 / 1M tokens (Magistral Medium) |
| Parameters | 116.8B total / 5.1B active (MoE) | Magistral Small: 24B (≈23.6B). Magistral Medium: undisclosed (built on Mistral Medium 3). |
| License | Apache 2.0 | Magistral Small: Apache 2.0 (open weights). Magistral Medium: proprietary, API-only. |
| Weights | Open weights | Open weightssame |
| Modalities | Text | Textsame |
| Knowledge cutoff | June 2024 | Not disclosed by Mistral. |
Benchmarks both models report
| Benchmark | gpt-oss-120b | Magistral 1.0 (Small & Medium) |
|---|---|---|
| GPQA Diamond | 80.1% | 70.8% |
| AIME 2024 | 96.6% | 73.6% |
| AIME 2025 | 97.9% | 64.9% |
Scores are as published by each maker or its cited source; harness, reasoning effort and date can differ between rows, so treat small gaps as ties. Sources are listed on each model page.
Which one should you pick?
Pick gpt-oss-120b if…
- You care about GPQA Diamond, AIME 2024, AIME 2025 (it scores higher on 3 shared benchmarks).
- Cost matters: input tokens cost 12.8× less ($0 vs $0.5 per 1M).
- You need the longer context (131K vs 128K).
- Your use case is: on-premise / private reasoning assistants where data cannot leave the org.
Pick Magistral 1.0 (Small & Medium) if…
- Your use case is: self-hosted, privacy-sensitive reasoning workloads using the open-weight Magistral Small.
FAQ
Which is better, gpt-oss-120b or Magistral 1.0 (Small & Medium)?
Gpt-oss-120b scores higher on 3 of the 3 benchmarks both models report; gpt-oss-120b is cheaper on input tokens ($0 vs $0.5 per 1M, 12.8×); gpt-oss-120b has the larger context window (131K vs 128K). The better pick depends on whether you value benchmark scores, price per token, context length or openness for your workload.
How much cheaper is gpt-oss-120b?
gpt-oss-120b: $0.039 / 1M tokens input, $0.18 / 1M tokens output. Magistral 1.0 (Small & Medium): $0.50 / 1M tokens (Magistral Small); $2.00 / 1M tokens (Magistral Medium) input, $1.50 / 1M tokens (Magistral Small); $5.00 / 1M tokens (Magistral Medium) output (list API prices as published by each maker).
What are the context windows of gpt-oss-120b and Magistral 1.0 (Small & Medium)?
gpt-oss-120b offers 131K with up to 131K output; Magistral 1.0 (Small & Medium) offers 128K with up to Not separately published; recommended working window is 40K tokens (performance may degrade beyond that, within the 128K maximum). output.
Is gpt-oss-120b newer than Magistral 1.0 (Small & Medium)?
gpt-oss-120b was released 2025-08-05; Magistral 1.0 (Small & Medium) was released 2025-06-10.