Comments (1)
This is expected. When you build from source it will use the CUDA toolkit installed on your system. At runtime it will try to link with a binary versioned with the same CUDA version from PyTorch by default. You can export BNB_CUDA_VERSION=124
to override that behavior and have it load libbitsandbytes_cuda124.so
instead.
from bitsandbytes.
Related Issues (20)
- dequantize_4bit() gives wrong output when working in cuda graph mode HOT 4
- Runtime Error, cannot import name 'get_keys_to_not_convert' from 'transformers.integrations' HOT 1
- Crash running FSDP on BF16-prequantized models HOT 4
- 4bit quantized model.dequantize() fails on CPU
- libbitsandbytes_cpu.so,libbitsandbytes_cuda124_nocublaslt124.so HOT 4
- libcudart.so Not Found HOT 2
- Unable to override PyTorch CUDA Version HOT 4
- Communicate blocksize constraints to kernels that take blocksize as a runtime argument HOT 6
- About fusion of **kdequantize kernel** and **simple bf16/fp16 matmul** HOT 1
- where are the outliers stored in LLM.int8 quantization for inference suing transformers library on AMD GPU? HOT 1
- 'nf4' compute datatype? HOT 1
- RuntimeError: CUDA Setup failed despite GPU being available. Please run the following command to get more information: HOT 1
- Error while trying to install the multi-backend-refactor branch for rocm in WSL2 HOT 1
- Error occurred when executing DownloadAndLoadFlorence2Model: HOT 1
- RuntimeError: Failed to import transformers.integrations.bitsandbytes because of the following error (look up to see its traceback): HOT 1
- Any plan to support block size 32? HOT 4
- Pretrained Causal LM cannot be loaded in 4bit/8bit HOT 6
- Linear8bitLt can not be moved back to cpu
- quantize_4bit/dequantize_4bit gives wrong output on in-contiguous tensor
- Cannot load decoder.lm_head.weight when loading 4 bit quantized model using VisionEncoderDecoder.from_pretrained HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from bitsandbytes.