how to replace the default LLaMA3 model with my fine-tuned LLaMA3?

[win4r]

If I follow the steps outlined here "https://github.com/phidatahq/phidata/tree/main/cookbook/llms/ollama/rag",
can I replace the default LLaMA3 model with my fine-tuned LLaMA3?

[ysolanky]

@win4r the tutorial you linked, demonstrates running llama3 using Ollama. You can use any model that's available on Ollama using Phidata by passing the model id as a param to the Ollama class

Example: assistant = Assistant(llm=Ollama(model="model_id"))

Can you share more details about your fine-tuned llama3 model? How are you hosting/running the model?

[chen-li1314]

@win4r the tutorial you linked, demonstrates running llama3 using Ollama. You can use any model that's available on Ollama using Phidata by passing the model id as a param to the Ollama class

Example: assistant = Assistant(llm=Ollama(model="model_id"))

Can you share more details about your fine-tuned llama3 model? How are you hosting/running the model?

After replacing the fine-tuning model, I received an error ResponseError: model 'llama3' not found, try pulling it first

[chen-li1314]

@win4r the tutorial you linked, demonstrates running llama3 using Ollama. You can use any model that's available on Ollama using Phidata by passing the model id as a param to the Ollama class
Example: assistant = Assistant(llm=Ollama(model="model_id"))
Can you share more details about your fine-tuned llama3 model? How are you hosting/running the model?

After replacing the fine-tuning model, I received an error ResponseError: model 'llama3' not found, try pulling it first


My local olama will report an error

What is API/embeddings
who can help me

[ju1987yetchung]

@win4r the tutorial you linked, demonstrates running llama3 using Ollama. You can use any model that's available on Ollama using Phidata by passing the model id as a param to the Ollama class
Example:
Can you share more details about your fine-tuned llama3 model? How are you hosting/running the model?assistant = Assistant(llm=Ollama(model="model_id"))

After replacing the fine-tuning model, I received an error ResponseError: model 'llama3' not found, try pulling it first


My local olama will report an error

What is API/embeddings
who can help me

I have same problem. mark this issue and wait an answer.