gpt-oss-20b-GGUF

Original Model

Run with LlamaEdge

• LlamaEdge version: v0.25.0 and above (0.25.1+ with tool call support)
• Prompt template
  • Prompt type: gpt-oss
  • Prompt string
    text

    1<|start|>system<|message|> 2You are ChatGPT, a large language model trained by OpenAI. 3Knowledge cutoff: 2024-06 4Current date: 2025-08-06 5Reasoning: medium 6# Valid channels: analysis, commentary, final. Channel must be included for every message. 7<|end|> 8 9<|start|>user<|message|>Hello!<|end|> 10<|start|>assistant<|channel|>final<|message|>Hi there!<|end|> 11<|start|>user<|message|>What's your favorite color?<|end|> 12<|start|>assistant
• Context size: 128000
• Run as LlamaEdge service
  bash

  1wasmedge --dir .:. --nn-preload default:GGML:AUTO:gpt-oss-20b-MXFP4_MOE.gguf \ 2 llama-api-server.wasm \ 3 --model-name gpt-oss-20b \ 4 --prompt-template gpt-oss \ 5 --ctx-size 128000