OpenAI’s open-weight models designed for powerful reasoning, agentic tasks
100K+
GGUF version by Unsloth
Welcome to the gpt-oss series, OpenAI’s open-weight models designed for powerful reasoning, agentic tasks, and versatile developer use cases.
| Model variant | Parameters | Quantization | Context window | VRAM¹ | Size |
|---|---|---|---|---|---|
ai/gpt-oss:latestai/gpt-oss:20B-UD-Q4_K_XL | 20B | MOSTLY_Q4_K_M | 131K tokens | 11.97 GiB | 11.04 GB |
ai/gpt-oss:20B-F16 | 20B | MOSTLY_F16 | 131K tokens | 13.25 GiB | 12.83 GB |
ai/gpt-oss:20B-UD-Q4_K_XL | 20B | MOSTLY_Q4_K_M | 131K tokens | 11.97 GiB | 11.04 GB |
ai/gpt-oss:20B-UD-Q6_K_XL | 20B | MOSTLY_Q6_K | 131K tokens | 12.12 GiB | 11.20 GB |
ai/gpt-oss:20B-UD-Q8_K_XL | 20B | MOSTLY_Q8_0 | 131K tokens | 12.69 GiB | 12.28 GB |
¹: VRAM estimated based on model characteristics.
latest→20B-UD-Q4_K_XL
Run the model:
docker model run ai/gpt-oss
Content type
Model
Digest
sha256:9398339cb…
Size
11.1 GB
Last updated
3 months ago
docker model pull ai/gpt-oss:20BPulls:
11,299
Last week