leejet
Bug can't use umt5 small for wan to save ram
[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.1.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.0.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model fi[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.0.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.1.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.0.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file |==================================> | 850/1223 - 2103.96it/s[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.2.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.0.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model fil[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] [INFO ] [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.1.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file el file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.3.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.0.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.1.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.4.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - nknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file 1][INFO ] [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model fil[INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.1.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.5.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.0.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.1.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.6.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.0.SelfAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.0.SelfAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.0.SelfAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.0.SelfAttention.relative_attention_bias.weight | f32 | 2 [6, 32, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.0.SelfAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.0.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.1.EncDecAttention.k.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.1.EncDecAttention.o.weight | f32 | 2 [384, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.1.EncDecAttention.q.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.1.EncDecAttention.v.weight | f32 | 2 [512, 384, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.1.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.2.DenseReluDense.wi_0.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.2.DenseReluDense.wi_1.weight | f32 | 2 [512, 1024, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.2.DenseReluDense.wo.weight | f32 | 2 [1024, 512, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.block.7.layer.2.layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [INFO ] model.cpp:2383 - unknown tensor 'text_encoders.t5xxl.transformer.decoder.final_layer_norm.weight | f32 | 1 [512, 1, 1, 1, 1]' in model file [ERROR] model.cpp:2392 - tensor 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.q.weight' has wrong shape in model file: got [512, 384, 1, 1], expected [4096, 4096, 1, 1] [INFO ] [ERROR] [INFO ] model.cpp:2392 - tensor 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight' has wrong shape in model file: got [6, 32, 1, 1], expected [64, 32, 1, 1] model.cpp:2392 - tensor 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight' has wrong shape in model file: got [6, 32, 1, 1], expected [64, 32, 1, 1] [WARN ] model.cpp:2392 - tensor 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.v.weight' has wrong shape in model file: got [512, 384, 1, 1], emodel.cpp:2392 - tensor 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.v.weight' has wrong shape in model file: got [512, 384, 1, 1], expected [4096, 4096, 1, 1] [ERROR] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.q.weight' [WARN ] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight' [ERROR] [WARN ] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight' model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.relative_attention_bias.weight' [WARN ] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.v.weight' model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.v.weight' tention_bias.weight' , 384, 1, 1], expected [4096, 4096, 1, 1] 64, 32, 1, 1] [ERROR] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.v.weight' [WARN ] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.o.weight' [INFO ] model.cpp:2179 - process tensor failed: 'text_encoders.t5xxl.transformer.encoder.block.0.layer.0.SelfAttention.o.weight' [INFO ] model.cpp:2351 - loading tensors completed, taking 0.64s (process: 0.04s, read: 0.32s, memcpy: 0.00s, convert: 0.00s, copy_to_backend: 0.00s) [ERROR] model.cpp:2408 - load tensors from file failed [ERROR] stable-diffusion.cpp:620 - load tensors from model loader failed new_sd_ctx_t failed
The wan video model uses umt5 xxl instead of umt5 small.
Oh ik I just was hoping you would add support for it it doesn't work in comfyui ether
Do they really embed into the same latent space? If not, this would not make much sense.
I may be mistaken, but I beleive umt5 small is a model trained on its own rather than a features-based distilled version of umt5 xxl, which means the embeddings generated by umt5 small would be completely unintelligible for wan models even if we add support for it.
