LLM
GA
Open weights
gpt-oss-20b
Local assistant on a 16 GB-class GPU
Release2025-08-05
Context131K
Max output32,768 tokens
Parameters21B
Local / VRAM12–16 GB
LicenseOpen weights
Overview
gpt-oss-20b is a model in the LLM category from OpenAI. It was released in 2025-08-05. The vendor is based in US. Best for: Local assistant on a 16 GB-class GPU.
Strengths
- 16 GB VRAM
- Apache-2.0
- Fast inference
Limits
- Not as strong as 120B
- No image input
Specifications
| Vendor | OpenAI |
|---|---|
| Category | LLM |
| Release | 2025-08-05 |
| Status | GA |
| License | Open weights |
| License name | Apache-2.0 |
| Commercial use | Allowed |
| Parameters | 21B |
| Active parameters | 3.6B |
| Architecture | MoE transformer |
| Context window | 131K |
| Max output tokens | 32,768 |
| Input modalities | Text |
| Output modalities | Text |
| Prompt types | Text |
| Negative prompts | No |
| Max outputs | 1 |
| Local deployment | Yes |
| VRAM | 12–16 GB |
| System RAM | 16–32 GB |
| API access | Yes |
| Pricing | Free weights |
| Confidence | High |
| Vendor country | US |
| Vendor type | Frontier lab |
Pricing & hardware
Free weights
This model can be self-hosted or run locally. Approximate GPU memory: 12–16 GB. Approximate system RAM: 16–32 GB. API access is available.