README.md

---
license: apache-2.0
library_name: peft
base_model: meta-llama/Llama-3.3-70B-Instruct
base_model_relation: adapter
pipeline_tag: text-generation
language:
  - en
datasets:
  - manifesta/brandvoice-marketing-briefs
tags:
  - lora
  - peft
  - marketing
  - brand-voice
  - copywriting
  - style-transfer
  - llama
  - autoscientist
  - adaption
inference: false
model-index:
  - name: brandvoice-marketing-model
    results:
      - task:
          type: text-generation
          name: Marketing copy generation
        dataset:
          name: BrandVoice Marketing Briefs (held-out)
          type: manifesta/brandvoice-marketing-briefs
          split: test
        metrics:
          - type: copy-quality
            name: Copy quality, Adaption judge (0 to 10)
            value: 8.7
          - type: improvement-percent
            name: Quality improvement vs base
            value: 24.3
          - type: win-rate
            name: Win rate vs base
            value: 56.0
          - type: percentile
            name: Quality percentile
            value: 30
        source:
          name: Adaption AutoScientist held-out evaluation
          url: https://huggingface.co/manifesta/brandvoice-marketing-model
---

<p align="center">
  <img src="https://huggingface.co/manifesta/brandvoice-marketing-model/resolve/main/assets/banner.png" width="100%" alt="BrandVoice: a brand-voice adapter for marketing copy">
</p>

[![Live demo](https://img.shields.io/badge/Live_demo-brand--voice--lab-000?logo=vercel&logoColor=white)](https://brand-voice-lab.vercel.app)
[![Base model](https://img.shields.io/badge/Base-Llama_3.3_70B-1B72E8?logo=meta&logoColor=white)](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct)
[![Dataset](https://img.shields.io/badge/%F0%9F%A4%97_Dataset-brandvoice--marketing--briefs-FFD21E)](https://huggingface.co/datasets/manifesta/brandvoice-marketing-briefs)
[![Quality](https://img.shields.io/badge/Copy_quality-%2B24.3%25-2ea44f)](#evaluation)
[![Win rate](https://img.shields.io/badge/Win_rate-56%25-2ea44f)](#evaluation)
[![Built with AutoScientist](https://img.shields.io/badge/Built_with-Adaption_AutoScientist-7C3AED)](https://adaptionlabs.ai)

# BrandVoice Marketing Model

A LoRA adapter that makes Llama 3.3 70B write marketing copy in a specific brand's voice instead of the flat, voiceless prose every model defaults to. Give it a brief and a brand-voice fingerprint, and it writes the headline, tagline, body, or CTA the way that brand actually sounds.

This is the piece most challenge entries skip. The dataset is public, the evaluation is public, and the trained weights are public too. You can load this adapter and reproduce the result yourself.

## Contents

- [TL;DR](#tldr)
- [Quickstart](#quickstart)
- [Evaluation](#evaluation)
- [Training](#training)
- [Intended use and limits](#intended-use-and-limits)
- [Dataset](#dataset)
- [License](#license)
- [Citation](#citation)

## TL;DR

- **What it is:** a PEFT LoRA adapter on `meta-llama/Llama-3.3-70B-Instruct`. Small, mergeable, cheap to serve.
- **What it does:** turns a marketing brief plus a 15-field brand-voice fingerprint into short on-brand copy.
- **The number:** copy quality went 7.0 to 8.7, a +24.3% relative lift, with a 56% win rate against the exact base it was trained on, scored by Adaption's held-out judge.
- **Built by:** MANIFESTA (Aivaras Navardauskas) for the Adaption AutoScientist Challenge, Marketing.

## Quickstart

Load the base model, then attach the adapter with PEFT.

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_id = "meta-llama/Llama-3.3-70B-Instruct"   # gated, request access first
adapter_id = "manifesta/brandvoice-marketing-model"

tokenizer = AutoTokenizer.from_pretrained(base_id)
base = AutoModelForCausalLM.from_pretrained(base_id, torch_dtype=torch.bfloat16, device_map="auto")
model = PeftModel.from_pretrained(base, adapter_id)
model.eval()

brief = "Write a launch tweet for our new oat-milk latte. Voice: dry, confident, a little funny."
messages = [{"role": "user", "content": brief}]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)
out = model.generate(inputs, max_new_tokens=200, temperature=0.7)
print(tokenizer.decode(out[0], skip_special_tokens=True))
```

One-line alternative that reads the base from `adapter_config.json`:

```python
from peft import AutoPeftModelForCausalLM
model = AutoPeftModelForCausalLM.from_pretrained("manifesta/brandvoice-marketing-model")
```

## Evaluation

Evaluated on a held-out set of marketing briefs the adapter never saw in training, scored by Adaption's AutoScientist judge. The base model is the same Llama 3.3 70B the adapter was trained on, so this is a clean before-and-after, not a comparison against a weaker model.

| Metric | Base | BrandVoice | Change |
| --- | --- | --- | --- |
| Copy quality (0 to 10) | 7.0 | 8.7 | +24.3% |
| Win rate vs base | 44% | 56% | +12 pts |
| Quality percentile | 8.4 | 30.0 | +21.6 |
| Eval loss | 1.543 | 1.375 | -0.168 |

Win rate is head-to-head preference on the same briefs, judged blind. The +24.3% is the relative lift in mean judge score (7.0 to 8.7). One judge, one held-out split, reported honestly. No leaderboard cherry-picking.

## Training

Adapted with Adaption AutoScientist (Adaptive Data). The platform synthesizes a fitting prompt for each completion, conditions on the 15-field voice fingerprint, and trains the adapter.

| Hyperparameter | Value |
| --- | --- |
| Base model | `meta-llama/Llama-3.3-70B-Instruct` |
| Method | LoRA (PEFT) |
| LoRA rank | 8 |
| LoRA alpha | 8 |
| LoRA dropout | 0.0 |
| Target modules | `q_proj`, `v_proj` |
| Learning rate | 3e-4 |
| Epochs | 1 |
| Precision | bfloat16 |
| Weights export | Together AI |

Recipe: `deduplication` and `prompt_rephrase` on, no `reasoning_traces` (copy is not chain-of-thought). A universal marketing blueprint rode on every completion: lead with the reader's benefit, be concrete, match the format and length, one clear call to action, never invent fake stats or claims.

Honest note: this is a first, light run. One epoch and a small rank-8 adapter. The lift is real and the full pipeline (ingest, adapt, evaluate, export, publish) is proven end to end. Bigger runs are headroom, not a different method. We would rather publish a true 24.3% than a dressed-up number you cannot reproduce.

## Intended use and limits

**Use it for**

- Generating on-brand marketing copy from a brief plus a voice fingerprint.
- Rewriting a rough draft into a target brand's voice (style transfer).
- Powering copywriting assistants and on-brand-ness tooling.

**Do not use it for**

- Passing generated text off as a brand's official communication.
- Deceptive, fabricated, or unsubstantiated advertising claims.
- Any high-stakes use (legal, medical, financial advice). This is a style model, not a source of truth.

**Limits**

- Voice fingerprints are inferred from public copy, not signed off by brands. A strong descriptive prior, not ground truth.
- English only. Best at short form (headlines, taglines, short body), weaker on long narrative.
- It shifts style, not the base model's knowledge or safety profile. Keep the base safety stack in place.
- Strongly stylized voices can pull outputs toward edgy phrasing. Bound it with the blueprint and safety controls in production.

## Dataset

Trained on [BrandVoice Marketing Briefs](https://huggingface.co/datasets/manifesta/brandvoice-marketing-briefs), released CC0. 6,403 lines of real marketing copy from 83 brands, each line carrying a 15-field voice fingerprint (tone, formality, humor, rhythm, personality, signature moves, do and don't rules, vocabulary, voice summary, format, length, category, task) at a 100% fill rate. A split of brands is held out for evaluation, so the reported gain is on unseen voices.

## License

- **Adapter weights:** Apache-2.0.
- **Base model:** governed by the [Llama 3.3 Community License](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct). You must comply with it to load and use the base.
- **Training dataset:** CC0-1.0.
- Brand names are referenced descriptively. Trademarks belong to their owners. Not affiliated with or endorsed by any brand named.

## Citation

```bibtex
@misc{brandvoice2026,
  title  = {BrandVoice: a brand-voice adapter and dataset for marketing copy},
  author = {Navardauskas, Aivaras},
  year   = {2026},
  note   = {MANIFESTA. Adapted with Adaptive Data by Adaption, AutoScientist Challenge, Marketing.},
  howpublished = {\url{https://huggingface.co/manifesta/brandvoice-marketing-model}}
}
```

Credit: Adaptive Data by Adaption. Dataset and adapter built by MANIFESTA (Aivaras Navardauskas) for the AutoScientist Challenge.