Update README.md

README.md

@@ -19,6 +19,20 @@ datasets:
 
 # Manacá-1B (base)
 
+<p align="center">
+  <img src="manacaidentity.png" width="340" alt="Manacá — Tibouchina mutabilis: os três estágios florais como metáfora do treinamento do LLM / the three flowering colours as a metaphor for language-model maturation"/>
+</p>
+
+<p align="center"><em>LLM base aberto e reprodutível para o português do Brasil · Open, reproducible Brazilian-Portuguese base LLM</em></p>
+
+<p align="center">
+  <a href="https://creativecommons.org/licenses/by/4.0/"><img src="https://img.shields.io/badge/License-CC%20BY%204.0-lightgrey.svg" alt="License: CC BY 4.0"></a>
+  <a href="https://github.com/Instituto-IA-LNCC/manaca-1b-base"><img src="https://img.shields.io/badge/Code-GitHub-181717.svg?logo=github" alt="Code"></a>
+  <img src="https://img.shields.io/badge/Language-PT--BR-009c3b.svg" alt="PT-BR">
+  <img src="https://img.shields.io/badge/Params-1.72B-purple.svg" alt="1.72B">
+  <img src="https://img.shields.io/badge/Base-pretrained-8A2BE2.svg" alt="base">
+</p>
+
 **[🇧🇷 Português](#português)** · **[🇬🇧 English](#english)**
 
 Manacá-1B é um modelo de linguagem decoder-only de ~1,72 bilhão de parâmetros,
@@ -85,6 +99,10 @@ warmup de 2.000), z-loss, bf16, *recompute* full, seed 1234. Treinado em 2 GPUs
 24 GB (sem NVLink) com paralelismo de dados (ZeRO-1). A corrida foi estável: 0
 iterações puladas e 0 NaN; loss de treino 11,41 → 2,48; loss de validação 2,07 nats.
 
+<p align="center">
+  <img src="training_dynamics_en.png" width="760" alt="Dinâmica de pré-treino do Manacá-1B: loss de treino e validação, norma do gradiente e learning rate ao longo de ~42B tokens"/>
+</p>
+
 ### Avaliação
 
 Acurácia (%) em quatro benchmarks de português, mesmo harness. CALAME-PT por geração
@@ -104,6 +122,10 @@ de acaso em ARC-Challenge-PT (como todo modelo base nessa escala). Detalhes, tes
 pareados de McNemar e validação do harness no
 [repositório](https://github.com/Instituto-IA-LNCC/manaca-1b-base) e no [preprint](https://github.com/Instituto-IA-LNCC/manaca-1b-base/blob/main/paper/manaca_1b_base_arxiv.pdf).
 
+<p align="center">
+  <img src="benchmarks_paper_en.png" width="640" alt="Manacá-1B em quatro benchmarks de português, acurácia vs. parâmetros (escala log), com IC95%"/>
+</p>
+
 ### Uso
 
 ```python
@@ -186,6 +208,10 @@ warmup), z-loss, bf16, full recompute, seed 1234. Trained on 2 GPUs of 24 GB (no
 NVLink) with data parallelism (ZeRO-1). The run was stable: 0 skipped and 0 NaN
 steps; training loss 11.41 → 2.48; validation loss 2.07 nats.
 
+<p align="center">
+  <img src="training_dynamics_en.png" width="760" alt="Manacá-1B pretraining dynamics: training and validation loss, gradient norm, and learning rate over ~42B tokens"/>
+</p>
+
 ### Evaluation
 
 Accuracy (%) on four Portuguese benchmarks, one harness. CALAME-PT by last-word
@@ -205,6 +231,10 @@ does every base model at this scale). Details, paired McNemar tests, and harness
 validation are in the
 [repository](https://github.com/Instituto-IA-LNCC/manaca-1b-base) and the [preprint](https://github.com/Instituto-IA-LNCC/manaca-1b-base/blob/main/paper/manaca_1b_base_arxiv.pdf).
 
+<p align="center">
+  <img src="benchmarks_paper_en.png" width="640" alt="Manacá-1B across four Portuguese benchmarks, accuracy vs. parameters (log scale), with 95% CIs"/>
+</p>
+
 ### Usage
 
 ```python
@@ -244,7 +274,7 @@ on multiple choice). Evaluate before any sensitive use.
             Porto, Fabio Andr\'e Machado},
   year   = {2026},
   note   = {LNCC (AI Institute) $\times$ NII/LLM-jp},
-  url    = {https://github.com/brunoleomenezes/manaca-1b-base}
+  url    = {https://github.com/Instituto-IA-LNCC/manaca-1b-base}
 }
 ```