Evren Yalçın

It seems to be fixed.

thank you for reply @rbren I'm not talking about performance. The Ollama doesn't work at all. I keep getting these errors.

``` ollama list NAME ID SIZE MODIFIED llama2:latest 78e26419b446 3.8 GB 2 days ago orca-mini:latest 2dbd9f439647 2.0 GB 2 days ago ``` ``` docker run \ -e LLM_API_KEY="ollama" \ -e...

@Umpire2018 ``` export LLM_MODEL="ollama/llama2" export LLM_API_KEY="ollama" export LLM_EMBEDDING_MODEL="local" export WORKSPACE_DIR="./workspace" export LLM_BASE_URL="http://localhost:11434" root@33803e5885c6:/app/opendevin# python main.py --task "write a bash script that prints hello" Running agent MonologueAgent (model: ollama/llama2, directory: None)...

Similar to what you said, I got a result like this in docker. @SmartManoj ``` import requests from litellm import completion messages = [ { "role": "user", "content": "Hello, how...

Thank you for reply @rbren A vulnerability that occurs when a user visits an attacker-controlled website. You can think of it as malicious JS code on the attacker's server. You...

@iFurySt https://github.com/OpenDevin/OpenDevin/issues/1194 At first it was about that, but with the use of **auth and TLS** the impact of this can be reduced. It's already been discussed here. #1248. I...

@enyst @rbren Could you please marked this as Security label

Thanks @enyst ^^

ref : https://github.com/OpenDevin/OpenDevin/issues/1208