Skip to content

Adjust Qwen Image quant type #370

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
YarvixPA wants to merge 1 commit into
base: auto_convert
Choose a base branch
from
Open

Adjust Qwen Image quant type #370

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 10 additions & 9 deletions tools/lcpp.patch
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -126,7 +126,7 @@ index 24e1f1f0..ee68edfd 100644
// get hparams kv
ml.get_key(LLM_KV_VOCAB_SIZE, hparams.n_vocab, false) || ml.get_arr_n(LLM_KV_TOKENIZER_LIST, hparams.n_vocab);

@@ -18016,6 +18072,158 @@ static void llama_tensor_dequantize_internal(
@@ -18016,6 +18072,159 @@ static void llama_tensor_dequantize_internal(
workers.clear();
}

Expand Down Expand Up @@ -237,14 +237,15 @@ index 24e1f1f0..ee68edfd 100644
+ (name.find("transformer_blocks.0.") != std::string::npos) ||
+ (name.find("transformer_blocks.59.") != std::string::npos) // this should be dynamic
+ ) {
+ if (ftype == LLAMA_FTYPE_MOSTLY_Q3_K_S || ftype == LLAMA_FTYPE_MOSTLY_Q2_K) {
+ new_type = GGML_TYPE_Q4_K;
+ }
+ else if (ftype == LLAMA_FTYPE_MOSTLY_Q3_K_M || ftype == LLAMA_FTYPE_MOSTLY_Q3_K_L) {
+ new_type = GGML_TYPE_Q4_K;
+ }
+ else if (ftype == LLAMA_FTYPE_MOSTLY_Q4_K_M) {
+ new_type = GGML_TYPE_Q5_K;
+ if (ftype == LLAMA_FTYPE_MOSTLY_Q2_K ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q3_K_S ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q3_K_M ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q3_K_L ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q4_0 ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q4_1 ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q4_K_S ||
+ ftype == LLAMA_FTYPE_MOSTLY_Q4_K_M) {
+ new_type = GGML_TYPE_Q5_K; // Minimum Q5_K for low quants
+ }
+ else if (ftype == LLAMA_FTYPE_MOSTLY_Q5_K_M) {
+ new_type = GGML_TYPE_Q6_K;
Expand Down