Mirror_Repos/ggml-org_llama.cpp
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

server : warn swa-full is not supported for non-SWA models (#20291)
Copilot Setup Steps / copilot-setup-steps (push) Has been cancelled
Details
Check Pre-Tokenizer Hashes / pre-tokenizer-hashes (push) Has been cancelled
Details
Python check requirements.txt / check-requirements (push) Has been cancelled
Details
Python Type-Check / pyright type-check (push) Has been cancelled
Details
Update Operations Documentation / update-ops-docs (push) Has been cancelled
Details

Browse Source
This commit is contained in:
Georgi Gerganov
2026-03-09 16:44:25 +02:00
committed by GitHub
parent d6e1556499
commit 344ee2a38a
1 changed files with 7 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-context.cpp
+7
View File
@@ -729,6 +729,13 @@ private:
}
}
if (llama_model_n_swa(model) == 0) {
if (params_base.swa_full) {
params_base.swa_full = false;
SRV_WRN("%s\n", "swa_full is not supported by this model, it will be disabled");
}
}
// Necessary similarity of prompt for slot selection
slot_prompt_similarity = params_base.slot_prompt_similarity;