[BUG]: Error when install dynamo

[sunzx8]

Describe the Bug

when i run command
pip install "ai-dynamo[all]"

it raise error:

Steps to Reproduce

run pip install "ai-dynamo[all]"

Expected Behavior

run pip install "ai-dynamo[all]"

Actual Behavior

run pip install "ai-dynamo[all]"

Environment

x86_64
TencentOS Server release 3.1 (Final)
NAME="TencentOS Server"
VERSION="3.1 (Final)"
ID="tencentos"
ID_LIKE="rhel fedora centos"
VERSION_ID="3.1"
PLATFORM_ID="platform:el8"
PRETTY_NAME="TencentOS Server 3.1 (Final)"
ANSI_COLOR="0;31"
CPE_NAME="cpe:/o:tencentos:tencentos:3"
HOME_URL="https://cloud.tencent.com/product/ts"

TENCENT_SUPPORT_PRODUCT="tencentos"
TENCENT_SUPPORT_PRODUCT_VERSION="3"
NAME_ORIG="TencentOS Server"

TencentOS Server release 3.1 (Final)
TencentOS Server 3.1 (Final)
TencentOS Server 3.1 (Final)
TencentOS Server 3.1 (Final)

Additional Context

No response

Screenshots

No response

[lauragonzalezmoran]

Happened to me too.
The problem was that The python3 -m venv method installs Python 3.13.2 by default on Ubuntu 24.04, which is not supported by ai-dynamo-runtime (https://pypi.org/project/ai-dynamo-runtime/) . To avoid dependency resolution issues, I used Conda to create a Python 3.12 environment, which is supported by ai-dynamo-runtime:

Step 1. Create conda env (if you don't have conda installed, you must do that first) :
conda create --name dynamo-test2 python=3.12
conda activate dynamo-test2

Step 2. Then Install the ai-dynamo[all] package:
pip install "ai-dynamo[all]"

With these changes I was able to do dynamo run out=vllm deepseek-ai/DeepSeek-R1-Distill-Llama-8B and chat with the model.

However, I have encountered further issues, which I have reported in #1101 .

[sunzx8]

Happened to me too. The problem was that The python3 -m venv method installs Python 3.13.2 by default on Ubuntu 24.04, which is not supported by ai-dynamo-runtime (https://pypi.org/project/ai-dynamo-runtime/) . To avoid dependency resolution issues, I used Conda to create a Python 3.12 environment, which is supported by ai-dynamo-runtime:

Step 1. Create conda env (if you don't have conda installed, you must do that first) : conda create --name dynamo-test2 python=3.12 conda activate dynamo-test2

Step 2. Then Install the ai-dynamo[all] package: pip install "ai-dynamo[all]"

With these changes I was able to do dynamo run out=vllm deepseek-ai/DeepSeek-R1-Distill-Llama-8B and chat with the model.

However, I have encountered further issues, which I have reported in #1101 .

Hello, thank you for your response. I have already tried using conda to create a Python 3.12 environment, but I'm still encountering the same issue. I suspect it might be due to a different version of Python 3.12 installed by conda. I then followed another issue you mentioned and saw that you were using version 3.12.3. However, even after I installed 3.12.3, the error still persists.

[grahamking]

We are converting Dynamo to be Python-version agnostic. The last step is #1016 . After that we can release a version that should work with any recent-ish Python.

[grahamking]

Fix landed, will be in 0.3.0.