llama.cpp v0.7.0 Runs Qwen3.8 on 128GB Mac/Spark
AI summary
llama.cpp has released v0.7.0, and the author provides the command to run it on a 128GB Mac/Spark: llama serve -hf ggml-org/Qwen3.8-Flash-Next-GGUF:IQ4_NL --spec-type draft-mtp --lazy-mode on.
Body
llama.cpp v0.7.0, 128GB Mac/Spark:
llama serve -hf ggml-org/Qwen3.8-Flash-Next-GGUF:IQ4_NL --spec-type draft-mtp --lazy-mode on