gpt-oss-20b-GGUF

Overview

GGUF quantization of OpenAI's gpt-oss-20b, provided by Unsloth. A 20B parameter text generation model compatible with llama.cpp.

Strengths

• HumanEval+: 81.7% (134/164)
• MBPP+: 92.6% (350/378)

Limitations

• BFCL (tool calling): 6.8% (400 problems)

Tested Hardware

Benchmarked by Infersec on:

• NVIDIA RTX 5090 (64 GB RAM, 32 GB VRAM, linux)