LWM

LWM copied to clipboard

Memory requirements

5

It would be worth to provide the measured memory requirements for inference Text Models at 32K, 128K,256K,512K and 1M tokens context window in both PyTorch and JAX.

loretoparisi

vision chat error

7

Hi, I'm trying to run `run_vision_chat.sh` but getting the following error: ``` (lwm) minyoung@claw2:~/Projects/LWM$ bash scripts/run_vision_chat.sh I0215 18:19:20.605390 140230836105600 xla_bridge.py:689] Unable to initialize backend 'rocm': NOT_FOUND: Could not find registered...

Minyoung1005

Please provide example uses of the scripts

4

Only got image with vision jax model to work, and even then had to remove the mesh_grid arg. Everything else has failed. E.g. needle fails like: ``` #! /bin/bash export...

pseudotensor

Do you have any plans on creating safetensors for the models?

jakebonk

Thank you for your contribution! Amazing performance! I just wonder the computational requirements for training such world models, e.g., how many GPUs and how long you need to train it?

HaoranZhuExplorer

License

1

Hi, Great work on LWM! I noticed the weights are licensed under the Apache license but derived from Llama 2, do both the Llama 2 license and the Apache license...

fakerybakery

Mistral

6

Hi, LWM is incredible! Any plans to release a Mistral version? Thanks!

fakerybakery

Running into issues for mac M1

4

I am trying to run the `run_sample_video.sh` file from the scripts folder. I am running into a lot of dependency issues when running this on a mac M1. Has anyone...

Pkoiralap

LWM-Chat in PyTorch

1

Hi, thanks for releasing the code. Looks pretty interesting! I noticed that the LWM-Chat (multimodal) model checkpoint is only released in Jax. It would be great if you could release...

j-min

Hello there! Architecture innovator here! Everything preceding my model seems very inefficient.

nathanturk2