All models
OpenAI
gpt-oss 120B
OpenAI's open-weight reasoning model. Ships natively in MXFP4 and fits on a single 80 GB GPU.
moereasoningagentstools
Overview
gpt-oss 120B has 117B parameters with 5.1B active per token. OpenAI positions it for production reasoning workloads that fit on a single 80 GB GPU such as an H100. The context window is 131,072 tokens.
Hardware fit
The ggml-org GGUF is the native MXFP4 build at 63.4 GB, which leaves headroom for context on an 80 GB card.
| Quantization | Weights |
|---|---|
| MXFP4 | 63.4 GB |
Sizes are the GGUF file sizes from ggml-org/gpt-oss-120b-GGUF. Budget extra VRAM for the KV cache, which grows with context size and parallel requests.
Deploying on Loadcore
- In the console, open Models and search for
ggml-org/gpt-oss-120b-GGUF. - Pick the quantization that fits the target server and start the download.
- Start an instance with the context size you need and GPU layers set to offload the full model.
- Point your tools at
http://your-server:9090/v1with your personal key.
Recommended settings
| Parameter | Value |
|---|---|
| temperature | 1.0 |
| top_p | 1.0 |
| top_k | 0 |
| repeat_penalty | 1.0 |
| min_p | 0.0 |
The model card does not list sampling values. Temperature 1.0 and top_p 1.0 below are the commonly used defaults for this model, so tune them for your workload.
Source: openai/gpt-oss-120b.