Issues
- 1
What is the future plan of this torchchat project?
#1114 opened by yanbing-j - 0
Cannot build ET Runner
#1107 opened by byjlw - 2
Support for quantized llm for smaller memory devices
#1024 opened by jhetuts - 1
Can't run ET Install
#1106 opened by byjlw - 0
should have the ability to output debug artifacts when exporting to a .pte file
#1101 opened by byjlw - 1
Cannot install ET
#1103 opened by byjlw - 0
HelloGitHub Badge
#1105 opened by 521xueweihan - 0
- 7
- 1
[Distributed Inference] running under .inference_mode() = assert Dtensor NotImplementedError: Operator aten.matmul.default does not have a sharding strategy registered.
#1087 opened by lessw2020 - 4
[Distributed Inference] moving stage.submod to non-fp32 (bf16, fp16) results in dtensor assert "self.mask_buffer.data is not None"
#1086 opened by lessw2020 - 2
download for 3.1 is broken
#1069 opened by byjlw - 0
Slow eval performance for .pte models
#1066 opened by vmpuri - 6
Streamlit Browser Issue
#1033 opened by nobelchowdary - 0
Slimming down torchchat: Replace replace_attention_with_custom_sdpa_attention() with ET's implementation
#1058 opened by Jack-Khuu - 3
support load model locally
#1040 opened by irasin - 4
Open AI API Maturity
#973 opened by Jack-Khuu - 3
- 7
ImportError: cannot import name 'int4_weight_only' from 'torchao.quantization.quant_api'
#1017 opened by sadimanna - 1
- 1
[Feature request] Langchain Support - Chat Model
#1009 opened by raymon-io - 4
How to deploy a new model by torchchat?
#1038 opened by liu8060 - 0
Dataclass Type Enforcement
#1025 opened by vmpuri - 0
Make quantization a first class feature
#1032 opened by byjlw - 0
Improve support for and documentation of custom models
#1041 opened by Jack-Khuu - 4
Can't install requirements when using Python-3.12
#963 opened by malfet - 2
- 2
Include final softmax in forward() to run on GPU when compiling w/ aoti or other accelerator on et
#1012 opened by sunshinesfbay - 4
- 3
AOTI/DSO model does not run in Linux
#996 opened by lhl - 4
[Raspbian] streamlit GUI interface does not work / no documentation how to install
#1001 opened by sunshinesfbay - 7
`scripts/build_native.sh et` errors out
#985 opened by sunshinesfbay - 1
torchchat有没有推理加速的功能,目前不支持qwen1.5,qwen2系列吗?
#1010 opened by yawzhe - 7
Crashes with internal assert while parsing options
#976 opened by malfet - 2
No attribute 'prompt' and 'num_samples'
#1003 opened by YeonwooSung - 3
No attribute Prompt
#1002 opened by bookandlover - 4
- 2
Weird model behaviour on Server/Browser: Looks like it's not using the template
#989 opened by akhilreddy0703 - 2
Android app crash with stories110m model
#977 opened by iceychris - 1
Leverage the HF cache for models
#992 opened by byjlw - 1
Could we request support for a smallish (~4-5B param) modern vision LLM? LLava-1.6 or Nanollava?
#988 opened by kinchahoy - 1
Android App Should Crash gracefully when the tokenizer in the aar doesn't match the model
#978 opened by Jack-Khuu - 0
Add an "Intro to torchchat" diagram to the README
#982 opened by Jack-Khuu - 0
- 0
Update CLI arg builders to check for only args that the subcommand uses: Export/Generate
#932 opened by Jack-Khuu - 1
- 1
- 0
- 0
- 3