RRepoGEO

REPOGEO REPORT · LITE

22-hours/cabrita

Default branch main · commit 1ec79989 · scanned 6/7/2026, 6:58:12 PM

GitHub: 559 stars · 69 forks

AI VISIBILITY SCORE
35 /100
Critical
Category recall
0 / 2
Not recommended in any query
Rule findings
1 pass · 1 warn · 0 fail
Objective metadata checks
AI knows your name
3 / 3
Direct prompts that named your repo
HOW TO READ THIS REPORT

Action plan is what to do next — copy-pasteable changes prioritized by impact. Category visibility is the real GEO test: when a user asks an AI a brand-free question that should surface 22-hours/cabrita, does the AI actually recommend you — or your competitors? Objective checks verify the metadata signals AI engines weight first. Self-mention check detects whether AI even knows you exist by name.

Action plan — copy-paste fixes

3 prioritized changes generated by gemini-2.5-flash. Mark items done after you ship the fix.

OVERALL DIRECTION
  • hightopics#1
    Add relevant topics to the repository

    Why:

    COPY-PASTE FIX
    llm, llama, finetuning, portuguese, nlp, instruction-following, peft, lora, generative-ai, machine-learning
  • highreadme#2
    Strengthen the README's opening sentence to clarify the project's core identity

    Why:

    CURRENT
    This repository is intended to share all the steps and resources that we used to finetune our version of LLaMA.
    COPY-PASTE FIX
    This repository provides **Cabrita, a finetuned instruction-following LLaMA model for Portuguese**, along with all the steps and resources used in its development.
  • mediumhomepage#3
    Add a project homepage URL

    Why:

    COPY-PASTE FIX
    https://huggingface.co/22-hours/cabrita (or your project's dedicated page)

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

Category visibility — the real GEO test

Brand-free queries asked to google/gemini-2.5-flash. Did AI recommend you, or someone else?

Same questions for every model — switch tabs to compare answers and rankings.

Recall
0 / 2
0% of queries surface 22-hours/cabrita
Avg rank
—
Lower is better. #1 = top recommendation.
Share of voice
0%
Of all named tools, what % are you?
Top rival
huggingface/peft
Recommended in 2 of 2 queries
COMPETITOR LEADERBOARD
  1. huggingface/peft · recommended 2×
  2. Mistral 7B Instruct · recommended 1×
  3. Hugging Face Transformers · recommended 1×
  4. peft · recommended 1×
  5. Llama 2 7B Chat · recommended 1×
  • CATEGORY QUERY
    How can I finetune an instruction-following large language model for Portuguese text?
    you: not recommended
    AI recommended (in order):
    1. Mistral 7B Instruct
    2. Hugging Face Transformers
    3. peft
    4. Llama 2 7B Chat
    5. Gemma 2B Instruct
    6. Open Llama
    7. Falcon 7B Instruct
    8. Alpaca-Portuguese
    9. OpenAssistant Conversations Dataset (OASST1)
    10. ShareGPT-Portuguese
    11. LoRA
    12. bitsandbytes
    13. accelerate

    AI recommended 13 alternatives but never named 22-hours/cabrita. This is the gap to close.

    Show full AI answer
  • CATEGORY QUERY
    What are the best practices for adapting open-source LLMs to European Portuguese datasets?
    you: not recommended
    AI recommended (in order):
    1. Common Crawl
    2. OSCAR (Open Super-large Crawled ALMAnaCH coRpus)
    3. Portuguese Web Corpus (PTWaC)
    4. EuroParl Parallel Corpus
    5. News Crawl Corpus
    6. spaCy (explosion/spaCy)
    7. Hugging Face Transformers Tokenizers library (huggingface/tokenizers)
    8. NLTK (Natural Language Toolkit) (nltk/nltk)
    9. XLM-RoBERTa (XLM-R)
    10. mBERT (Multilingual BERT)
    11. BLOOM
    12. Llama 2
    13. Mistral 7B
    14. Mixtral 8x7B
    15. Hugging Face Transformers library (huggingface/transformers)
    16. LoRA (Low-Rank Adaptation) (huggingface/peft)
    17. QLoRA (Quantized LoRA) (huggingface/peft)
    18. DeepSpeed (microsoft/DeepSpeed)
    19. FSDP (Fully Sharded Data Parallel)
    20. Hugging Face Evaluate library (huggingface/evaluate)
    21. Portuguese GLUE (P-GLUE)
    22. ONNX Runtime (microsoft/onnxruntime)
    23. TensorRT
    24. Hugging Face Inference Endpoints
    25. TGI (Text Generation Inference) (huggingface/text-generation-inference)
    26. MLflow (mlflow/mlflow)

    AI recommended 26 alternatives but never named 22-hours/cabrita. This is the gap to close.

    Show full AI answer

Objective checks

Rule-based audits of metadata signals AI engines weight most.

  • Metadata completeness
    warn

    Suggestion:

  • README presence
    pass

Self-mention check

Does AI even know your repo exists when asked about it directly?

  • Compared to common alternatives in this category, what is the core differentiator of 22-hours/cabrita?
    pass
    AI named 22-hours/cabrita explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • If a team adopts 22-hours/cabrita in production, what risks or prerequisites should they evaluate first?
    pass
    AI named 22-hours/cabrita explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

  • In one sentence, what problem does the repo 22-hours/cabrita solve, and who is the primary audience?
    pass
    AI named 22-hours/cabrita explicitly

    AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

Embed your GEO score

Drop this badge into the README of 22-hours/cabrita. It auto-updates whenever the report is rescanned and links back to the latest report — easy public proof that you care about AI discoverability.

RepoGEO badge previewLive preview
MARKDOWN (README)
[![RepoGEO](https://repogeo.com/badge/22-hours/cabrita.svg)](https://repogeo.com/en/r/22-hours/cabrita)
HTML
<a href="https://repogeo.com/en/r/22-hours/cabrita"><img src="https://repogeo.com/badge/22-hours/cabrita.svg" alt="RepoGEO" /></a>
Pro

Subscribe to Pro for deep diagnoses

22-hours/cabrita — Lite scans stay free; this card itemizes Pro deep limits vs Lite.

  • Deep reports10 / month
  • Brand-free category queries5 vs 2 in Lite
  • Prioritized action items8 vs 3 in Lite