Update README.md

README.md

@@ -3,7 +3,7 @@ language:
 - pt
 - en
 license: apache-2.0
-base_model: unsloth/qwen3-8b-bnb-4bit
+base_model: unsloth/qwen3-4b
 base_model_relation: finetune
 library_name: transformers
 pipeline_tag: text-generation
@@ -21,18 +21,18 @@ tags:
 - bitsandbytes
 ---
 
-# Jade8b
+# Jade4b
 
-Jade8b is a Brazilian Portuguese conversational finetune of Qwen3 8b built to express a strong, persistent persona. This model is designed for PT-BR chat, chatbot use cases, and character-style interaction, with colloquial language, abbreviations, slang, and a WhatsApp-like tone.
+Jade4b is a Brazilian Portuguese conversational finetune of Qwen3 4b built to express a strong, persistent persona. This model is designed for PT-BR chat, chatbot use cases, and character-style interaction, with colloquial language, abbreviations, slang, and a WhatsApp-like tone.
 
 ## Model Summary
 
-Jade8b is a persona-first model. It was intentionally finetuned so the model speaks like **Jade** even without a strong `system prompt`. Because of that, the model often answers in PT-BR with informal phrasing such as `vc`, slang, and a friendly conversational tone from the very first turn.
+Jade4b is a persona-first model. It was intentionally finetuned so the model speaks like **Jade** even without a strong `system prompt`. Because of that, the model often answers in PT-BR with informal phrasing such as `vc`, slang, and a friendly conversational tone from the very first turn.
 
 ## Model Details
 
 - Developed by: `Madras1`
-- Base model: `unsloth/qwen3-8b-bnb-4bit`
+- Base model: `unsloth/qwen3-4b`
 - Model type: conversational text-generation finetune
 - Primary language: Brazilian Portuguese (`pt-BR`)
 - License: `apache-2.0`
@@ -112,7 +112,7 @@ Example prompts:
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
 
-model_id = "Madras1/Jade8b"
+model_id = "Madras1/Jade4b"
 
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 model = AutoModelForCausalLM.from_pretrained(