Shushi Hong

[TFLite][Frontend] Support quantized Reverse sequence

1

Support Reverse sequence quantization operation as part of #15148

[TFLite][Frontend] Support quantized ARG_MIN

3

Support ARG_MIN quantization operation as part of #15148

[Model] Support Minicpm

2

This pr support Minicpm-2B, MOE version of Minicpm will be updated in following prs. Demonstration here: ``` tlopex@tlopex-OMEN-by-HP-Laptop-17-ck1xxx:~/mlc-llm$ mlc_llm chat dist/MiniCPM-2B-128k-q4f16_1-MLC --device "cuda:0" --overrides context_window_size=2048 --model-lib ./dist/libs/MiniCPM-2B-128k-q4f16_1-cuda.so [2024-08-06 00:01:19] INFO...

[Model] Add use_qk_norm option for Cohere model

3

This pr updates `use_qk_norm` option for Cohere series models like Command-R-Plus.

[CI] Move conda config files to tests/conda and remove unused conda build infrastructure

3

## Summary This PR reorganizes the conda-related files and removes unused conda build infrastructure. ## Changes ### Moved Files - `conda/build-environment.yaml` → `tests/conda/build-environment.yaml` - `conda/condarc` → `tests/conda/condarc` ### Removed Files...