Upload medpsy-v3.pte with huggingface_hub

Add Qwen3 1.7B official tokenizer config

Add Qwen3 1.7B official tokenizer (json)

Add Qwen3 1.7B official 8da4w model for fallback testing

Fix: add get_max_context_len metadata (root cause of iOS Failed to load LLM runner)

remove v6 (missing get_max_context_len metadata)

Fix: v6 export with 8da4w + 4w embedding (was 8w) — attempt to resolve iOS 'Failed to load LLM runner'

remove old v5.1 broken version (8da4w+8w)

Rename model.pte -> medpsy-v3.pte (Fix get_eos_ids plural metadata)

remove temporary model.pte (will be created as medpsy-v3.pte)

remove old version (no get_eos_ids plural)

Fix: add get_eos_ids (plural) metadata for react-native-executorch runtime

Upload medpsy-v3.pte with huggingface_hub

Upload README.md with huggingface_hub

Upload tokenizer_config.json with huggingface_hub

Upload tokenizer.json with huggingface_hub

initial commit