mtmd, common: various fixes (#27071)
* apply fixes * cont * revert gguf fix
This commit is contained in:
@@ -4077,6 +4077,9 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
|
||||
{"--spec-draft-n-max"}, "N",
|
||||
string_format("number of tokens to draft for speculative decoding (default: %d)", params.speculative.draft.n_max),
|
||||
[](common_params & params, int value) {
|
||||
if (value < 0) {
|
||||
throw std::invalid_argument("invalid value");
|
||||
}
|
||||
params.speculative.draft.n_max = value;
|
||||
}
|
||||
).set_spec().set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_LOOKUP, LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_CLI}).set_env("LLAMA_ARG_SPEC_DRAFT_N_MAX"));
|
||||
|
||||
Reference in New Issue
Block a user