Upload MLX packaging for OpenMed-ZeroShot-NER-DNA-Multi-209M-mlx

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,94 @@
+---
+license: apache-2.0
+base_model: OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M
+pipeline_tag: token-classification
+library_name: openmed
+tags:
+  - openmed
+  - mlx
+  - apple-silicon
+  - zero-shot-ner
+  - gliner
+  - medical
+  - clinical
+---
+
+# OpenMed-ZeroShot-NER-DNA-Multi-209M for OpenMed MLX
+
+This repository contains an OpenMed MLX conversion of [`OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M`](https://huggingface.co/OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M) for Apple Silicon inference with [OpenMed](https://github.com/maziyarpanahi/openmed).
+
+Artifact metadata:
+
+- OpenMed MLX task: `zero-shot-ner`
+- OpenMed MLX family: `gliner-uni-encoder-span`
+- Weight format: `safetensors`
+- Runtime API: `GLiNERMLXPipeline`
+
+[OpenMed](https://github.com/maziyarpanahi/openmed) is the main product experience:
+
+- Install the Python package with `pip install openmed`
+- Enable Apple Silicon acceleration with `pip install "openmed[mlx]"`
+- Load this MLX model directly from the Hub or from a local snapshot
+- For Apple apps, use OpenMedKit from the same GitHub repository with a compatible CoreML bundle
+
+This MLX repo is meant to pair with:
+
+- OpenMed GitHub: [https://github.com/maziyarpanahi/openmed](https://github.com/maziyarpanahi/openmed)
+- OpenMed website: [https://openmed.life](https://openmed.life)
+- Source checkpoint: [`OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M`](https://huggingface.co/OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M)
+
+## Quick Start
+
+```bash
+pip install openmed
+pip install "openmed[mlx]"
+```
+
+```python
+from huggingface_hub import snapshot_download
+from openmed.mlx.inference import GLiNERMLXPipeline
+
+model_path = snapshot_download("OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M-mlx")
+pipe = GLiNERMLXPipeline(model_path)
+
+entities = pipe.predict_entities(
+    "Patient John Doe was seen at Stanford Hospital.",
+    labels=["person", "organization", "location"],
+    threshold=0.5,
+)
+
+for entity in entities:
+    print(entity)
+```
+
+Prompt packing metadata included with the model:
+
+```json
+{
+  "kind": "gliner-words",
+  "entity_token": "<<ENT>>",
+  "separator_token": "<<SEP>>",
+  "class_token_index": 250103,
+  "embed_marker_token": true,
+  "split_mode": "words"
+}
+```
+
+## Swift and Apple Apps
+
+Use Swift with OpenMedKit, not with MLX weight files directly.
+
+1. Open Xcode and go to File > Add Package Dependencies.
+2. Paste the OpenMed repository URL: `https://github.com/maziyarpanahi/openmed`
+3. Choose the package product OpenMedKit from the repository.
+4. Add a compatible CoreML model bundle plus `id2label.json` to your app target.
+
+This MLX model is for Python services on Apple Silicon, local MLX inference on macOS, and Hub-hosted model distribution. If a given environment cannot write `weights.safetensors`, OpenMed falls back to `weights.npz` so the model remains usable.
+
+## Credits
+
+- Base checkpoint: [`OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M`](https://huggingface.co/OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M)
+- OpenMed GitHub: [https://github.com/maziyarpanahi/openmed](https://github.com/maziyarpanahi/openmed)
+- OpenMed website: [https://openmed.life](https://openmed.life)
+- MLX conversion and runtime support: OpenMed
+- Swift runtime for Apple apps: OpenMedKit from the OpenMed repository

config.json

@@ -0,0 +1,87 @@
+{
+  "transformers_version": "5.5.0",
+  "architectures": [
+    "GLiNERSpanModel"
+  ],
+  "output_hidden_states": false,
+  "return_dict": true,
+  "dtype": null,
+  "chunk_size_feed_forward": 0,
+  "is_encoder_decoder": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1"
+  },
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1
+  },
+  "problem_type": null,
+  "vocab_size": 250105,
+  "hidden_size": 512,
+  "num_hidden_layers": 12,
+  "num_attention_heads": 12,
+  "intermediate_size": 3072,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "attention_probs_dropout_prob": 0.1,
+  "max_position_embeddings": 512,
+  "type_vocab_size": 0,
+  "initializer_range": 0.02,
+  "layer_norm_eps": 1e-07,
+  "relative_attention": true,
+  "max_relative_positions": -1,
+  "pad_token_id": 0,
+  "bos_token_id": null,
+  "eos_token_id": null,
+  "position_biased_input": false,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "pooler_dropout": 0.0,
+  "pooler_hidden_act": "gelu",
+  "legacy": true,
+  "tie_word_embeddings": true,
+  "pooler_hidden_size": 768,
+  "_name_or_path": "OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M",
+  "model_type": "deberta-v2",
+  "position_buckets": 256,
+  "norm_rel_ebd": "layer_norm",
+  "share_att_key": true,
+  "output_attentions": false,
+  "encoder_hidden_size": 768,
+  "dropout": 0.4,
+  "max_width": 12,
+  "num_rnn_layers": 1,
+  "embed_ent_token": true,
+  "embed_rel_token": true,
+  "model_name": "microsoft/mdeberta-v3-base",
+  "ent_token": "<<ENT>>",
+  "sep_token": "<<SEP>>",
+  "_mlx_task": "zero-shot-ner",
+  "_mlx_family": "gliner-uni-encoder-span",
+  "_mlx_model_type": "deberta-v2",
+  "_mlx_runtime": {
+    "experimental": true
+  },
+  "class_token_index": 250103,
+  "_mlx_prompt_spec": {
+    "kind": "gliner-words",
+    "entity_token": "<<ENT>>",
+    "separator_token": "<<SEP>>",
+    "class_token_index": 250103,
+    "embed_marker_token": true,
+    "split_mode": "words"
+  },
+  "embed_class_token": true,
+  "rel_token_index": null,
+  "text_token_index": null,
+  "example_token_index": null,
+  "pooling_strategy": "first",
+  "class_token_pooling": "first",
+  "logit_scale_init_value": 1.0,
+  "_mlx_position_offset": 0,
+  "embedding_size": 768,
+  "_mlx_weights_format": "safetensors"
+}

id2label.json

@@ -0,0 +1,4 @@
+{
+  "0": "LABEL_0",
+  "1": "LABEL_1"
+}

openmed-mlx.json

@@ -0,0 +1,37 @@
+{
+  "format": "openmed-mlx",
+  "format_version": 2,
+  "task": "zero-shot-ner",
+  "family": "gliner-uni-encoder-span",
+  "source_model_id": "OpenMed/OpenMed-ZeroShot-NER-DNA-Multi-209M",
+  "config_path": "config.json",
+  "label_map_path": "id2label.json",
+  "preferred_weights": "weights.safetensors",
+  "fallback_weights": [
+    "weights.npz"
+  ],
+  "available_weights": [
+    "weights.safetensors"
+  ],
+  "weights_format": "safetensors",
+  "quantization": null,
+  "max_sequence_length": 512,
+  "tokenizer": {
+    "path": ".",
+    "files": [
+      "tokenizer.json",
+      "tokenizer_config.json"
+    ]
+  },
+  "prompt_spec": {
+    "kind": "gliner-words",
+    "entity_token": "<<ENT>>",
+    "separator_token": "<<SEP>>",
+    "class_token_index": 250103,
+    "embed_marker_token": true,
+    "split_mode": "words"
+  },
+  "runtime": {
+    "experimental": true
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4429a446347c64e61dae3f01f56cb59b4ac4afa78b72dd04cd9a28475f122202
+size 16034568

tokenizer_config.json

@@ -0,0 +1,123 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "extra_special_tokens": [
+    "[PAD]",
+    "[CLS]",
+    "[SEP]",
+    "▁<extra_id_99>",
+    "▁<extra_id_98>",
+    "▁<extra_id_97>",
+    "▁<extra_id_96>",
+    "▁<extra_id_95>",
+    "▁<extra_id_94>",
+    "▁<extra_id_93>",
+    "▁<extra_id_92>",
+    "▁<extra_id_91>",
+    "▁<extra_id_90>",
+    "▁<extra_id_89>",
+    "▁<extra_id_88>",
+    "▁<extra_id_87>",
+    "▁<extra_id_86>",
+    "▁<extra_id_85>",
+    "▁<extra_id_84>",
+    "▁<extra_id_83>",
+    "▁<extra_id_82>",
+    "▁<extra_id_81>",
+    "▁<extra_id_80>",
+    "▁<extra_id_79>",
+    "▁<extra_id_78>",
+    "▁<extra_id_77>",
+    "▁<extra_id_76>",
+    "▁<extra_id_75>",
+    "▁<extra_id_74>",
+    "▁<extra_id_73>",
+    "▁<extra_id_72>",
+    "▁<extra_id_71>",
+    "▁<extra_id_70>",
+    "▁<extra_id_69>",
+    "▁<extra_id_68>",
+    "▁<extra_id_67>",
+    "▁<extra_id_66>",
+    "▁<extra_id_65>",
+    "▁<extra_id_64>",
+    "▁<extra_id_63>",
+    "▁<extra_id_62>",
+    "▁<extra_id_61>",
+    "▁<extra_id_60>",
+    "▁<extra_id_59>",
+    "▁<extra_id_58>",
+    "▁<extra_id_57>",
+    "▁<extra_id_56>",
+    "▁<extra_id_55>",
+    "▁<extra_id_54>",
+    "▁<extra_id_53>",
+    "▁<extra_id_52>",
+    "▁<extra_id_51>",
+    "▁<extra_id_50>",
+    "▁<extra_id_49>",
+    "▁<extra_id_48>",
+    "▁<extra_id_47>",
+    "▁<extra_id_46>",
+    "▁<extra_id_45>",
+    "▁<extra_id_44>",
+    "▁<extra_id_43>",
+    "▁<extra_id_42>",
+    "▁<extra_id_41>",
+    "▁<extra_id_40>",
+    "▁<extra_id_39>",
+    "▁<extra_id_38>",
+    "▁<extra_id_37>",
+    "▁<extra_id_36>",
+    "▁<extra_id_35>",
+    "▁<extra_id_34>",
+    "▁<extra_id_33>",
+    "▁<extra_id_32>",
+    "▁<extra_id_31>",
+    "▁<extra_id_30>",
+    "▁<extra_id_29>",
+    "▁<extra_id_28>",
+    "▁<extra_id_27>",
+    "▁<extra_id_26>",
+    "▁<extra_id_25>",
+    "▁<extra_id_24>",
+    "▁<extra_id_23>",
+    "▁<extra_id_22>",
+    "▁<extra_id_21>",
+    "▁<extra_id_20>",
+    "▁<extra_id_19>",
+    "▁<extra_id_18>",
+    "▁<extra_id_17>",
+    "▁<extra_id_16>",
+    "▁<extra_id_15>",
+    "▁<extra_id_14>",
+    "▁<extra_id_13>",
+    "▁<extra_id_12>",
+    "▁<extra_id_11>",
+    "▁<extra_id_10>",
+    "▁<extra_id_9>",
+    "▁<extra_id_8>",
+    "▁<extra_id_7>",
+    "▁<extra_id_6>",
+    "▁<extra_id_5>",
+    "▁<extra_id_4>",
+    "▁<extra_id_3>",
+    "▁<extra_id_2>",
+    "▁<extra_id_1>",
+    "▁<extra_id_0>"
+  ],
+  "is_local": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "split_by_punct": false,
+  "tokenizer_class": "DebertaV2Tokenizer",
+  "unk_id": 3,
+  "unk_token": "[UNK]",
+  "vocab_type": "spm"
+}

weights.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb02a569688a3ce1320f98e5025ca54beb1241d59608c04475cfbe3ee7b5634
+size 1155815268