Model card: multimodal embeddings now run on GPU (LAST pooling end-state); re-ingest note

GPU multimodal embeddings: Gemma 4 12B native end-state path (LAST pooling + patched llamafile). image/audio embed on GPU; --search-file works on GPU. Verified on 2x RTX 4090.

Model card: correct multimodal-embedding framing (Gemma 4 12B native interleave; dedicated graph is the stopgap)

Model card: clarify GPU multimodal behavior (media indexes via text bridge; raw vectors auto-skipped on GPU)

Fix GPU multimodal ingest: orchestrator skips raw-media-embedding on GPU (was 0 chunks for any media corpus); media now indexes via text bridge. Verified on 2x RTX 4090.

Model card: GPU usage section + hardware-verified note (2x RTX 4090, CUDA 12.8)

GPU-capable build: orchestrator honors --gpu (-ngl offload), verified end-to-end on 2x RTX 4090 (CUDA 12.8); includes macOS embedded-payload extraction fix

Upload chimera-full.ape with huggingface_hub

Upload chimera-full.ape with huggingface_hub

Upload README.md with huggingface_hub

initial commit