runanywhere/cosmos3_edge_HNPU

Add v79 (SM8750/Hexagon v79) bundle (#1)

card: describe the model type (vision-language model — reads/answers) + capabilities

card: minimal public-safe form (benchmark-free, per convention)

card: cross-link the image repo

model card: public-ready details + on-device benchmarks (v81)

add text (chat) manifest: this VL repo now serves chat + vision from the shared decoder (understanding model)

Upload v81/vis_f16.bin with huggingface_hub

Upload v81/tokenizer.json with huggingface_hub

Upload v81/cosmos3edge_text_lmh_f16.bin with huggingface_hub

Upload v81/cosmos3edge_text_embed_f16.bin with huggingface_hub

Upload v81/cosmos3edge_text_dec_w8.bin with huggingface_hub

Upload v81/cosmos3-edge-vlm.json with huggingface_hub

Upload README.md with huggingface_hub

initial commit