sleepy/llama.cpp
1
0
Fork 0
Code Issues 12 Pull Requests Actions 113 Packages Projects Releases Wiki Activity

fit-params : refactor + add option to output estimated memory per device (#22171)

Browse Source 
* fit-params : add option to output estimated memory per device

* cont : minor

* cont : refactor

* cont : move fit params implementation to libcommon

* cont : header

* cont : headers

* cont : codeowners
This commit is contained in:
Georgi Gerganov
2026-04-21 09:54:36 +03:00
committed by GitHub
parent ff6b1062af
commit cfe9838d26
19 changed files with 1123 additions and 980 deletions
Download Patch File Download Diff File Expand all files Collapse all files
common/arg.cpp
+14
View File
@@ -2426,6 +2426,20 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
}
).set_env("LLAMA_ARG_FIT"));
add_opt(common_arg(
{ "-fitp", "--fit-print" }, "[on|off]",
string_format("print the estimated required memory ('on' or 'off', default: '%s')", params.fit_params_print ? "on" : "off"),
[](common_params & params, const std::string & value) {
if (is_truthy(value)) {
params.fit_params_print = true;
} else if (is_falsey(value)) {
params.fit_params_print = false;
} else {
throw std::runtime_error(
string_format("error: unknown value for --fit-print: '%s'\n", value.c_str()));
}
}
).set_examples({LLAMA_EXAMPLE_FIT_PARAMS}).set_env("LLAMA_ARG_FIT_ESTIMATE"));
add_opt(common_arg(
{ "-fitt", "--fit-target" }, "MiB0,MiB1,MiB2,...",
string_format("target margin per device for --fit, comma-separated list of values, "