redfox@homeserver:~/llama.cpp/build/bin$ ./llama-server -m “/home/redfox/llama.cpp/models/qwen3/Qwen3.5-9B-UD-Q5_K_XL.gguf?download=true” -t 22 -ngl 99 --host 0.0.0.0 --port 8888 --no-mmap --spec-type draft-mtp --spec-draft-n-max 4 ggml_cuda_init: found 1 CUDA devices (Total VRAM: 8109 MiB): Device 0: NVIDIA P104-100, compute capability 6.1, VMM: yes, VRAM: 8109 MiB error while handling argument “–spec-type”: unknown speculative decoding type without draft model
usage: –spec-type [none|ngram-cache|ngram-simple|ngram-map-k|ngram-map-k4v|ngram-mod] type of speculative decoding to use when no draft model is provided (default: none)
Не работает:
redfox@homeserver:~/llama.cpp/build/bin$ ./llama-server -m “/home/redfox/llama.cpp/models/qwen3/Qwen3.5-9B-UD-Q5_K_XL.gguf?download=true” -t 22 -ngl 99 --host 0.0.0.0 --port 8888 --no-mmap --spec-type draft-mtp --spec-draft-n-max 4 ggml_cuda_init: found 1 CUDA devices (Total VRAM: 8109 MiB): Device 0: NVIDIA P104-100, compute capability 6.1, VMM: yes, VRAM: 8109 MiB error while handling argument “–spec-type”: unknown speculative decoding type without draft model
usage: –spec-type [none|ngram-cache|ngram-simple|ngram-map-k|ngram-map-k4v|ngram-mod] type of speculative decoding to use when no draft model is provided (default: none)
(env: LLAMA_ARG_SPEC_TYPE)
to show complete usage, run with -h
Сейчас чтобы подключится к некоторым VPN сервисам необходимо в свою очередь использовать VPN чтобы открыть сайт сервиса.