Remove the "100% local" sidebar badge

Enlarge nav icons, icon the board, align top controls

Accent the user-facing next-action labels

Make the Q&A system prompt accent-correct PT-BR

Teach the Q&A model the agency quote/order/booking vocabulary

Add accents and cedillas to the PT-BR UI strings

Add Limpar/Clear button to start over

Fix identity demo holder name (matched transcription)

Log Stage-2 extraction + fix demo name leakage + demo toggle

Default back to fine-tunes; preload the transformers checkpoints

Log which model serves each inference, not just at load

Default vision + text to base models

Make QA GGUF env-overridable + fix stale preload list

Default to published v1 vision + v4 text models

Consolidate vision finetune files under finetune/vision/

Wire v3 train, merge, GGUF export, and runtime selection

Format dspy_extractor after glossary extraction

Re-accent extraction glossary and prompts (shared constant)

Use fp16 for the torch transcriber on pre-Ampere GPUs

Cap image size for the llama.cpp vision slicer

Prefer the torch transcriber over CTranslate2 on the GPU backend

Gate embedder bf16 on GPU compute capability, not is_bf16_supported

Shrink QA context to fit the T4 alongside other models

Use fp16 for the embedder on pre-Ampere GPUs

Run faster-whisper on GPU, never CPU

Bake the live Space's production models as defaults

Drop transformers-v46 backend; route cuda runtime to llama.cpp

Show honest counts on Resumo and translate presenter chrome

Answer Q&A in the selected UI language (EN default)

Add bilingual i18n translation table (default English)

Align MiniCPM5 v1 QA grounding behavior

Expose a persisted-work counter on every upload status tick

Render Q&A answers as Markdown in the chat dock

Add Nemotron Embed VL adapter and wire it per backend

Answer dashboard questions with hybrid retrieved context

Index the Q&A corpus as a final pipeline stage

Add deterministic MockEmbedder for offline retrieval tests

Store corpus chunks with embeddings in both repositories

Add Embedder port, corpus chunker, and hybrid retrieval logic

Show the thinking trace greyed out under a Pensando label

Stream Q&A phases through the dashboard interactor

Expose phased streaming on the TextModel port

Tell the Q&A model the asker is the travel agent

Recognize the agent by alias and rename her at ingest

Raise Q&A max output tokens to 2048

Enable MiniCPM5 thinking for Q&A by default

Support prompt-embedded think tags in reasoning strippers

Switch Q&A text model to MiniCPM5-1B-PTBR-v3

Build the hero desk objects (skeuomorphic tier 2)

Offer the synthetic example from the hero dropzone