Why do some models yield incomplete results? Such as Deepseek, Llama3.3:70b, etc

Why do some models yield incomplete results? Such as Deepseek, Llama3.3:70b, etc

`ruselt = client.chat(
                model="llama3.3:70b",
                # model="llama3.2-vision:11b",
                # model="gemma3:12b",
                # model=model,#
                prompt=content,
            )`
incomplete results