exo

exo copied to clipboard

samiamjidkhan

[WIP] Minimal Tokenizer Implementation

1

# Overview In response to #262 and #263, and building off of #271, I've been working on a minimal tokenizer for exo. Initially aimed to remove the `transformers` dependency, but...

snico432

Request: primary node or mix of manual/udp

1

Hi Great project, i really like how fast things are moving here! We have the following scenario: Multiple servers with GPUs in the same network, but they are "unstable". Some...

mr-deamon

Nodes do not find each other

12

I have 2 Ubuntu 22.04 Nodes, both started but do not find each other. Both nodes are on the same network. I have opened firewall port 52415 for tcp and...

schwaa

Request: respect `https_proxy` and `http_proxy` env variables for network requests

1

### Description I encountered an issue where `exo` does not respect the `https_proxy` and `http_proxy` environment variables for network requests, even though the system environment is properly configured. The root...

wanliqun

chore: support proxy environment variables in network requests

2

- Add `--trust-env` command line argument to enable aiohttp to respect `http_proxy` and `https_proxy` environment variables using monkey patching

wanliqun

None of PyTorch, TensorFlow >= 2.0, or Flax have been found. Models won't be available and only tokenizers, configuration and file/data utilities can be used.

3

None of PyTorch, TensorFlow >= 2.0, or Flax have been found. Models won't be available and only tokenizers, configuration and file/data utilities can be used. Installed exo on my two...

Mfespinosa92

Thunderbolt 5 mesh of Mac Minis

5

Hi there, I've just got 3 M4 Pro Minis to experiment with MLX, exo, etc. One thing I noticed that when I connect all of them in a ring: ```mermaid...

fkorotkov

How to load a model from a file

4

Hello! Thank you for your excellent work! This is the first distributed large model inference framework that I've seen that supports heterogeneity in a P2P way. I was interested and...

qtyandhasee

Feature request: support the qwq:32b model

1

Since Qwen models are in the list, I was wondering whether it is possible to support QwQ as well? Thanks!

0x7CFE