Four bugs were found in the Cloud-Edge Collaborative Inference for LLM example when running the benchmarks locally. Writing it all in this single issue because all of them are import errors.

1. Missing sedna dependency - core framework not installed automatically
2. Missing colorlog dependency in core requirements.txt
3. Missing retry dependency in example requirements.txt
4. Dead code import LadeSpecDecLLM causing ImportError

Bug #1: Missing sedna Framework

Description

Core Ianvs framework depends on sedna but it's not installed automatically via requirements.txt. Users must manually install from a wheel file in the repository.

Steps to Reproduce

git clone https://github.com/kubeedge/ianvs.git
cd ianvs
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install -e .

ianvs -f examples/cloud-edge-collaborative-inference-for-llm/benchmarkingjob.yaml

Error Output

Bug #2: Missing colorlog Dependency

Description

The core logging module imports colorlog but it's not listed in requirements.txt, causing failure on any benchmark run.

Steps to Reproduce

git clone https://github.com/kubeedge/ianvs.git
cd ianvs
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install -e .

ianvs -f examples/cloud-edge-collaborative-inference-for-llm/benchmarkingjob.yaml

Error Output

Bug #3: Missing retry Dependency

Description

The API-based LLM model imports retry decorator but it's not in the example requirements, causing crash when using cloud models.

Steps to Reproduce

# After fixing Bugs #1 and #2 and Run from ianvs root directory (not from example subdirectory)
source .venv/bin/activate

cd examples/cloud-edge-collaborative-inference-for-llm
pip install -r requirements.txt
cd ../.. 

export GROQ_API_KEY="your_key"
export GROQ_BASE_URL="https://api.groq.com/openai/v1"

ianvs -f examples/cloud-edge-collaborative-inference-for-llm/benchmarkingjob.yaml

Error Output

Bug #4: Dead Code Import LadeSpecDecLLM

Description

edge_model.py imports LadeSpecDecLLM which doesn't exist, causing ImportError on startup.

Steps to Reproduce

# After fixing Bugs #1, #2, and #3
source .venv/bin/activate

ianvs -f examples/cloud-edge-collaborative-inference-for-llm/benchmarkingjob.yaml

Error Output