Process.hierarchical forces openAI

[debankadas]

This works with local Ollama

crew = Crew( agents=[research_agent,summarize_agent], tasks=[task,task1], process=Process.sequential, memory=False, max_rpm=2, share_crew=False, manager_llm=mistral, function_calling_llm=mistral, verbose=2 )

This doesn't

crew = Crew( agents=[research_agent, summarize_agent], tasks=[task, task1], manager_llm=mistral, # Mandatory for hierarchical process process=Process.hierarchical, # Specifies the hierarchical management approach memory=True, max_rpm=2, function_calling_llm=mistral, verbose=2 )

The agents and task code remains same, so looks like Process.hierarchical is forcing openAI ?

[theCyberTech]

You havent specified the manager_llm attribute in your crew - https://docs.crewai.com/core-concepts/Processes/

[punitchauhan771]

crew = Crew(
agents=[research_agent, summarize_agent],
tasks=[task, task1],
manager_llm=mistral, # Mandatory for hierarchical process
process=Process.hierarchical, # Specifies the hierarchical management approach
memory=True,
max_rpm=2,
function_calling_llm=mistral,
verbose=2
)

mark memory = False, (when you use memory = True, it creates a rag and stores your data, for storing your data into a rag, it needs an embedding which by default is set to openai.) or you can provide your own embedder

embedder = dict(
            provider = "google",
            config = dict(
                model = 'models/embedding-001'
            )
        )
        
Crew(
            agents=[improver,expect],
            tasks=[improve_prompt_task, answer_question_task],
            process=Process.hierarchical,
            manager_llm=mistral,
            embedder = embedder,
            memory=True,
            function_calling_llm=mistral
        )

[joaomdmoura]

Exactly, but you can also set it to use other providers for memory too :): https://docs.crewai.com/core-concepts/Memory/#using-google-ai-embeddings

-- João Moura @joaomdmoura http://twitter.com/joaomdmoura

Em sex., 12 de abr. de 2024 às 01:56, punitchauhan771 < @.***> escreveu:

crew = Crew( agents=[research_agent, summarize_agent], tasks=[task, task1], manager_llm=mistral, # Mandatory for hierarchical process process=Process.hierarchical, # Specifies the hierarchical management approach memory=True, max_rpm=2, function_calling_llm=mistral, verbose=2 )

mark memory = False, (when you use memory = True, it creates a rag and stores your data, for storing your data into a rag, it needs a embedding which by default is set to openai.

— Reply to this email directly, view it on GitHub https://github.com/joaomdmoura/crewAI/issues/462#issuecomment-2050973998, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAFC3N7REAGOV56QEXELVXDY45SOPAVCNFSM6AAAAABGCQXSTKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJQHE3TGOJZHA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[github-actions[bot]]

This issue is stale because it has been open for 30 days with no activity. Remove stale label or comment or this will be closed in 5 days.

[github-actions[bot]]

This issue was closed because it has been stalled for 5 days with no activity.