[Reclamação] Mensagem ao rodar

Ao rodar o comando "embeddings.embed_query("Política de home office da empresa")" dará essa mensagem, pois excedemos os limites do uso gratuito do gemini, vocês poderiam fazer um exemplo que não acontecesse essa rate limit.

Erro: "ResourceExhausted Traceback (most recent call last)
/usr/local/lib/python3.12/dist-packages/langchain_google_genai/embeddings.py in embed_query(self, text, task_type, title, output_dimensionality)
323 )
--> 324 result: EmbedContentResponse = self.client.embed_content(request)
325 except Exception as e:

9 frames
ResourceExhausted: 429 You exceeded your current quota, please check your plan and billing details. For more information on this error, head to: https://ai.google.dev/gemini-api/docs/rate-limits.

Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0
Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0
Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0
Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0 [violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerMinutePerProjectPerModel-FreeTier"
}
violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerMinutePerUserPerProjectPerModel-FreeTier"
}
violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerDayPerUserPerProjectPerModel-FreeTier"
}
violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerDayPerProjectPerModel-FreeTier"
}
, links {
description: "Learn more about Gemini API quotas"
url: "https://ai.google.dev/gemini-api/docs/rate-limits"
}
]

The above exception was the direct cause of the following exception:

GoogleGenerativeAIError Traceback (most recent call last)
/usr/local/lib/python3.12/dist-packages/langchain_google_genai/embeddings.py in embed_query(self, text, task_type, title, output_dimensionality)
325 except Exception as e:
326 msg = f"Error embedding content: {e}"
--> 327 raise GoogleGenerativeAIError(msg) from e
328 return list(result.embedding.values)
329

GoogleGenerativeAIError: Error embedding content: 429 You exceeded your current quota, please check your plan and billing details. For more information on this error, head to: https://ai.google.dev/gemini-api/docs/rate-limits.

Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0
Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0
Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0
Quota exceeded for metric: generativelanguage.googleapis.com/embed_content_free_tier_requests, limit: 0 [violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerMinutePerProjectPerModel-FreeTier"
}
violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerMinutePerUserPerProjectPerModel-FreeTier"
}
violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerDayPerUserPerProjectPerModel-FreeTier"
}
violations {
quota_metric: "generativelanguage.googleapis.com/embed_content_free_tier_requests"
quota_id: "EmbedContentRequestsPerDayPerProjectPerModel-FreeTier"
}
, links {
description: "Learn more about Gemini API quotas"
url: "https://ai.google.dev/gemini-api/docs/rate-limits"
}
]
"

Oi, João Pedro! Como vai?

O erro que apareceu indica que o limite gratuito da API do Gemini foi atingido. Isso é comum quando usamos o plano free.

A forma mais simples de resolver é usar um modelo local gratuito no lugar do Gemini. Assim, você continua praticando sem depender da cota da API.

Veja este exemplo rápido:


from langchain_community.embeddings import HuggingFaceEmbeddings

# usa um modelo gratuito e leve
embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")

query = "Politica de home office da empresa"
print(embeddings.embed_query(query)[:5])

Esse código faz o mesmo tipo de embedding, só que sem precisar de API externa.

Espero ter ajudado. Conte com o apoio do Fórum na sua jornada. Fico à disposição.

Abraços e bons estudos!

Caso este post tenha lhe ajudado, por favor, marcar como solucionado