minicpm5-1b-vivamais-v4/README.md

language, license, base_model, datasets, library_name, pipeline_tag, tags, model-index

language	license	base_model	datasets	library_name	pipeline_tag	tags	model-index
pt	apache-2.0	marinarosa/minicpm5-1b-vivamais-v1	marinarosa/minicpm5-vivamais-text-sft-v4	transformers	text-generation	brazilian-portuguese minicpm5 vivamais travel-agency grounded-qa conversational distilled qlora lora redacted	name results minicpm5-1b-vivamais-v4 task dataset metrics type name text-generation Viva Mais dashboard QA name type split Viva Mais QA eval v4 marinarosa/minicpm5-vivamais-text-sft-v4 eval type value name average_score 0.5569620253164557 Average score type value name pass_rate 0.44936708860759494 Pass rate task dataset metrics type name text-generation Generic PT-BR conversational QA name type split Viva Mais generic PT-BR eval marinarosa/minicpm5-vivamais-text-sft-v4 eval type value name average_score 0.627906976744186 Average score type value name pass_rate 0.627906976744186 Pass rate

MiniCPM5-1B Viva Mais v4

marinarosa/minicpm5-1b-vivamais-v4 is a Brazilian Portuguese text model for Viva Mais, a local-first WhatsApp copilot for travel-agency CRM workflows. It is intended for grounded Q&A over extracted customer, payment, reservation, ticket, and document context.

This v4 candidate was published because it improves the aggregate dashboard and generic PT-BR eval scores over v1. It still has an important limitation: it often over-generates and may include adjacent customer facts after the direct answer. Use short generation limits and strong retrieval/context isolation in production.

Lineage

• Upstream family: openbmb/MiniCPM5-1B.
• Fine-tuning base for this release: marinarosa/minicpm5-1b-vivamais-v1.
• Published Transformers model: marinarosa/minicpm5-1b-vivamais-v4.
• GGUF export: marinarosa/minicpm5-1b-vivamais-v4-GGUF.
• Training dataset: marinarosa/minicpm5-vivamais-text-sft-v4.

Fine-Tuning Recipe

The v4 run used response-only supervised fine-tuning with a QLoRA-style training setup, followed by a merged 16-bit export.

• Hardware: A100 on Modal.
• Context length: 8192 tokens.
• Training rows: 4000.
• Epochs: 2.0.
• Steps: 500.
• Learning rate: 1e-05 with cosine schedule and 3% warmup.
• Per-device batch size: 2.
• Gradient accumulation: 8.
• Effective batch size: 16.
• Final training loss: 0.2231.
• Base loaded in 4-bit for training; final artifact saved as merged 16-bit safetensors.
• LoRA rank: 16.
• LoRA alpha: 32.
• LoRA dropout: 0.05.
• LoRA target modules: q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj.
• Response-only masking used <|im_start|>user and a no-thinking assistant prefix so the model trains only on final assistant answers.

Dataset Mix

Bucket	Rows
Grounding negatives / unknown-answer cases	2381
Viva Mais domain QA	1511
Rio 3.1 teacher-distilled chat	80
Generic PT-BR conversation remainder	28

Teacher distillation used prefeitura-rio/Rio-3.1-Open-4B-Instruct. The run accepted 80 rows and rejected 5 rows after filters for verbose wrappers, hidden-reasoning artifacts, weak travel claims, and category mismatch.

Evaluation

Run name: v4_rio_001. Baseline: marinarosa/minicpm5-1b-vivamais-v1. Candidate: marinarosa/minicpm5-1b-vivamais-v4.

Suite	Model	Avg score	Pass rate	Count
Viva Mais dashboard QA	v1 baseline	0.5042	0.4051	158
Viva Mais dashboard QA	v4 candidate	0.5570	0.4494	158
Generic PT-BR conversational QA	v1 baseline	0.5814	0.5814	43
Generic PT-BR conversational QA	v4 candidate	0.6279	0.6279	43

Dashboard QA details:

Metric	v1 baseline	v4 candidate
Average score	0.5042	0.5570
Pass rate	0.4051	0.4494
Leakage failures	54	55
Unknown failures	18	11
Internal gate passed	false	false

Viva Mais category scores for v4:

Category	Score
conversation_grounding	0.6667
dates	0.7222
document_status	0.7500
entity_isolation	0.8333
long_context_distractor	0.6250
not_enough_information	0.0556
outstanding_balance	0.5417
passenger_flight_lookup	0.4677
payment_status	0.6000
pipeline_stage	0.7500
quote_status	0.7500
what_is_missing	0.2500

Generic PT-BR category scores for v4:

Category	Score
customer_service	0.6250
entity_isolation	0.4000
grounded_refusal	0.8333
payment_math	0.6667
summary_rewrite	0.5714
travel_qa	0.3333
whatsapp_style	1.0000

Usage

Use the MiniCPM chat template and keep generation short for grounded CRM answers. For local llama.cpp inference, prefer the GGUF repo linked above.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "marinarosa/minicpm5-1b-vivamais-v4"
tok = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
    trust_remote_code=True,
)
messages = [
    {"role": "system", "content": "Responda em PT-BR usando apenas o contexto fornecido."},
    {"role": "user", "content": "Contexto: ...\nPergunta: qual saldo ainda falta?"},
]
text = tok.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
    enable_thinking=False,
)

Privacy and Safety

The training data is synthetic or redacted. No raw WhatsApp exports, identity documents, client names, CPF values, phone numbers, emails, card numbers, or local private data are published. This model is not a general travel advisor or payment authority; it should answer only from supplied context and defer when the context is insufficient.