params.speculative.cpuparams.n_threads = std::thread::hardware_concurrency();
}
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE}));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"-tbd", "--threads-batch-draft"}, "N",
"number of threads to use during batch and prompt processing (default: same as --threads-draft)",
params.speculative.cpuparams_batch.n_threads = std::thread::hardware_concurrency();
}
}
- ).set_examples({LLAMA_EXAMPLE_SPECULATIVE}));
+ ).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"-Cd", "--cpu-mask-draft"}, "M",
"Draft model CPU affinity mask. Complements cpu-range-draft (default: same as --cpu-mask)",
params_dft.cache_type_k = params_base.speculative.cache_type_k;
params_dft.cache_type_v = params_base.speculative.cache_type_v;
+ params_dft.cpuparams.n_threads = params_base.speculative.cpuparams.n_threads;
+ params_dft.cpuparams_batch.n_threads = params_base.speculative.cpuparams_batch.n_threads;
params_dft.tensor_buft_overrides = params_base.speculative.tensor_buft_overrides;
llama_init_dft = common_init_from_params(params_dft);