AutoGPT: In --gpt3only mode, receive this error: SYSTEM: Command evaluate_code returned: Error: The model: `gpt-4` does not exist

To disable gpt-4 just edit the configuration, set SMART_LLM_MODEL="gpt-3.5-turbo" in .env.

Having paid OpenAI account doesn’t neccessarily mean you have access to GPT4 (neither does ChatGPT Plus subscription), currently the only way to get access to GPT4 is through waitlist: https://openai.com/waitlist/gpt-4-api

Same. I have access to GPT4, but still get this error whenever evaluate_code is run.

@ljleb looks like it was fixed, no more gpt4 errors with the flag

PINECONE_API_KEY=your-pinecone-api-key PINECONE_ENV=your-pinecone-region OPENAI_API_KEY= ELEVENLABS_API_KEY= SMART_LLM_MODEL=“gpt-3.5-turbo-0301” FAST_LLM_MODEL=“gpt-3.5-turbo” GOOGLE_API_KEY= CUSTOM_SEARCH_ENGINE_ID= USE_AZURE=False OPENAI_API_BASE=your-base-url-for-azure OPENAI_API_VERSION=api-version-for-azure OPENAI_DEPLOYMENT_ID=deployment-id-for-azure IMAGE_PROVIDER=dalle HUGGINGFACE_API_TOKEN=

@ljleb 's response is correct - there are 2 models referenced in .env. I only had one set to 3.5.