philipwan
philipwan
* Update input from initializer if missing * add a test and comments * remove useless modification * fix flake8 * do not update shape if exisit in input *...
According to your method, I compiled tensorflow_aarch64 on tf2.5, but it reported an error: `ERROR: /root/tensorflow/tensorflow/python/BUILD:2531:29: Linking of rule '//tensorflow/python:gen_candidate_sampling_ops_py_wrappers_cc' failed (Exit 1): gcc failed: error executing command /usr/bin/gcc @bazel-out/host/bin/tensorflow/python/gen_candidate_sampling_ops_py_wrappers_cc-2.params...
# 🐛 Bug RuntimeError: unsupported output type: int, from operator: xformers::efficient_attention_forward_cutlass ## Command ## To Reproduce StableDiffusionXL model use with torch.jit and enable_xformers_memory_efficient_attention function ``` m.enable_xformers_memory_efficient_attention torch.jit.trace(m) ``` Steps to...
According to examples file and implementation of **pipeline.py**, every time using new prompt, you need to warm up at least **denoising_steps_num** times. Am I understanding it wrong? If different prompts...
Can Customers use FM module for inference without training？
open-sora-plan update to v1.2 version, model struct changed from spatial + temporal to (T*H*W) transformer block; It has a huge sequence_length, This inference cost is unacceptable，Can **PAB** provide additional support?...
如题，想了解下在新的结构中模型运行效果如何
In addition to flux, does nunchaku support prompt-to-video models, such as OpenSoraPlan, Cogvideo，etc. Flux is also a transformer-based model
When I calling flashinfer.single_prefill_with_kv_cache with param custom_mask, the error is ``` (VllmWorker rank=1 engine_index=0 pid=3101376) ERROR 10-26 11:09:15 [logger.py:146] File "/usr/local/lib/python3.12/dist-packages/flashinfer/prefill.py", line 1162, in single_prefill_with_kv_cache (VllmWorker rank=1 engine_index=0 pid=3101376) ERROR...
SD3 also has MMDiT module, can TeaCache support this model and bring small loss of accuracy