GPTOSS-120B-HauhauCS
No changes to datasets or capabilities. Fully functional, 100% of what the original authors intended.
Format
MXFP4 GGUF. This is the model's native precision - GPT-OSS was trained in MXFP4, so no further quantization is needed or recommended. Re-quantizing would only lose quality.
Works with llama.cpp, LM Studio, Ollama, and anything else that loads GGUFs.
Downloads
| File | Size |
|---|---|
| GPTOSS-120B-Uncensored-HauhauCS-Aggressive-MXFP4.gguf | 61 GB |
Specs
- 117B total parameters, ~5.1B active per forward pass (MoE: 128 experts, top-4 routing)
- 128K context
Recommended Settings
temperature: 1.0top_k: 40- Everything else (top_p, min_p, repeat penalty, etc.) should be disabled - some clients enable these by default, turn them off
Fits in ~61GB VRAM. Single H100 or equivalent. For lower VRAM, use --n-cpu-moe N in llama.cpp to offload MoE layers to CPU.