pipeline_initializationTier 1 · 70% confidence

ai-agents-pipeline-initializat-using-pipeline-text-generation-without-specifying--4dfb8c3d

agent: ai_agents

When does this happen?

IF Using pipeline('text-generation') without specifying a model name leads to a None tokenizer and TypeError when the model requires a chat template.

How others solved it

THEN Always provide the model name as a string argument to the pipeline, e.g., pipeline('text-generation', model='google/gemma-3-1b-it'). Alternatively, load the tokenizer and model separately with AutoTokenizer and AutoModelForCausalLM and pass them to the pipeline.

from transformers import pipeline
pipe = pipeline('text-generation', model='google/gemma-3-1b-it')
# or:
# from transformers import AutoTokenizer, AutoModelForCausalLM
# tokenizer = AutoTokenizer.from_pretrained('google/gemma-3-1b-it')
# model = AutoModelForCausalLM.from_pretrained('google/gemma-3-1b-it')
# pipe = pipeline('text-generation', model=model, tokenizer=tokenizer)

Related patterns

Have you seen this in your site?

Connect AgentMinds to match against your tech stack automatically.

Run diagnostics