vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

common : add -hfd option for the draft model (#11318)

Browse source 
* common : add -hfd option for the draft model

* cont : fix env var

* cont : more fixes
This commit is contained in:
Georgi Gerganov 2025-01-20 22:29:43 +02:00 • committed by GitHub
parent aea8ddd516
commit 80d0d6b4b7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 24 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

8
common/common.h
View file

@ -175,7 +175,11 @@ struct common_params_speculative {
struct cpu_params cpuparams;
struct cpu_params cpuparams_batch;
std::string model = ""; // draft model for speculative decoding // NOLINT
std::string hf_repo = ""; // HF repo // NOLINT
std::string hf_file = ""; // HF file // NOLINT
std::string model = ""; // draft model for speculative decoding // NOLINT
std::string model_url = ""; // model url to download // NOLINT
};
struct common_params_vocoder {
@ -508,12 +512,14 @@ struct llama_model * common_load_model_from_url(
const std::string & local_path,
const std::string & hf_token,
const struct llama_model_params & params);
struct llama_model * common_load_model_from_hf(
const std::string & repo,
const std::string & remote_path,
const std::string & local_path,
const std::string & hf_token,
const struct llama_model_params & params);
std::pair<std::string, std::string> common_get_hf_file(
const std::string & hf_repo_with_tag,
const std::string & hf_token);