model : allow Qwen3.8 Flash Next to accept image embds for MTP (#30397)
Cont of #30257, closes #30307
This commit is contained in:
1 parent
882d67166d
commit
7f51953d7d
1 file changed
+1
-1
@@ -526,7 +526,7 @@ llama_model_qwen4exp::graph::graph(const llama_model & model, const llm_graph_pa
|
||||
llama_model_qwen4exp::graph_mtp::graph_mtp(const llama_model & model, const llm_graph_params & params) :
|
||||
graph(model, params, no_build{}) {
|
||||
GGML_ASSERT(hparams.n_layer_nextn == 1 && "qwen4exp MTP has a single block");
|
||||
GGML_ASSERT(ubatch.token && "qwen4exp MTP requires token input");
|
||||
ASSERT_EMBD_OR_TOKEN(ubatch);
|
||||
|
||||
const int64_t hc = hparams.dsv4_hc_mult;
|
||||
GGML_ASSERT(hparams.n_embd_out() == (uint32_t) (n_embd*hc) && "qwen4exp MTP hidden width mismatch");
|
||||
|
||||
Reference in new issue
Block a user