openjev NLI workflow on ZeroGPU: bound jev-classify/jev-rerank fn nodes, requirements, gradio 6.28.0

README.md

@@ -4,7 +4,7 @@ emoji: ⚡
 colorFrom: indigo
 colorTo: indigo
 sdk: gradio
-sdk_version: 6.25.0
+sdk_version: 6.28.0
 app_file: run.py
 pinned: false
 hf_oauth: true

requirements.txt

@@ -0,0 +1,3 @@
+spaces
+torch
+transformers

run.py

@@ -1,6 +1,55 @@
+import json
+
 import gradio as gr
+import spaces
+import torch
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+
+MODEL_ID = "AlexWortega/openjev"
+SUBFOLDER = "qwen3.5-4b-nli-v2"
+LABELS = ["contradiction", "entailment", "neutral"]
+
+# Load at module level on cuda — ZeroGPU emulates CUDA at startup and
+# attaches a real GPU inside @spaces.GPU functions.
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, subfolder=SUBFOLDER)
+model = AutoModelForSequenceClassification.from_pretrained(
+    MODEL_ID, subfolder=SUBFOLDER, trust_remote_code=True, torch_dtype=torch.bfloat16
+).to("cuda").eval()
+
+
+@spaces.GPU(duration=60)
+def classify(premise: str, hypothesis: str) -> str:
+    """Run openjev NLI: returns contradiction / entailment / neutral probabilities."""
+    text = model.config.nli_template.format(premise=premise, hypothesis=hypothesis)
+    inputs = tokenizer(text, return_tensors="pt").to("cuda")
+    with torch.no_grad():
+        probs = model(**inputs).logits.softmax(-1)[0].float().cpu()
+    result = {label: round(float(p), 4) for label, p in zip(LABELS, probs)}
+    result["prediction"] = LABELS[int(probs.argmax())]
+    return json.dumps(result, indent=2)
+
+
+@spaces.GPU(duration=60)
+def rerank(question: str, options: str) -> str:
+    """Pick the option with the highest entailment against the question.
+    Options are one per line."""
+    best_idx, best_score, scores = -1, -1.0, []
+    for i, option in enumerate([o.strip() for o in options.splitlines() if o.strip()]):
+        text = model.config.nli_template.format(premise=question, hypothesis=option)
+        inputs = tokenizer(text, return_tensors="pt").to("cuda")
+        with torch.no_grad():
+            probs = model(**inputs).logits.softmax(-1)[0].float().cpu()
+        score = float(probs[1])  # entailment
+        scores.append((option, round(score, 4)))
+        if score > best_score:
+            best_idx, best_score = i, score
+    return json.dumps({"answer_index": best_idx, "scores": scores}, indent=2)
+
 
-demo = gr.Workflow()
+demo = gr.Workflow(
+    graph="workflow.json",
+    bind={"jev-classify": classify, "jev-rerank": rerank},
+)
 
 if __name__ == "__main__":
     demo.launch()

workflow.json

@@ -1 +1,74 @@
-{"schema_version":"2","name":"My Workflow","runtime":{"default":"client"},"references":[],"operators":[],"subjects":[],"edges":[],"view":{"default":"canvas"}}
+{
+  "schema_version": "2",
+  "name": "openjev NLI",
+  "runtime": {"default": "client"},
+  "references": [
+    {
+      "id": "ref_premise",
+      "label": "Premise",
+      "role": "reference",
+      "asset_type": "text",
+      "data": {"value": "The bird is 0.05 below the centre of the gap."},
+      "inputs": [{"id": "in", "label": "Text", "type": "text"}],
+      "outputs": [{"id": "out", "label": "Text", "type": "text"}],
+      "x": 60, "y": 80
+    },
+    {
+      "id": "ref_hypothesis",
+      "label": "Hypothesis",
+      "role": "reference",
+      "asset_type": "text",
+      "data": {"value": "The bird is below the centre of the gap."},
+      "inputs": [{"id": "in", "label": "Text", "type": "text"}],
+      "outputs": [{"id": "out", "label": "Text", "type": "text"}],
+      "x": 60, "y": 320
+    }
+  ],
+  "operators": [
+    {
+      "id": "op_classify",
+      "label": "jev-classify",
+      "role": "operator",
+      "kind": "fn",
+      "fn": "jev-classify",
+      "inputs": [
+        {"id": "premise", "label": "Premise", "type": "text", "required": true},
+        {"id": "hypothesis", "label": "Hypothesis", "type": "text", "required": true}
+      ],
+      "outputs": [{"id": "out_0", "label": "Result", "type": "text", "output_index": 0}],
+      "x": 460, "y": 180
+    }
+  ],
+  "subjects": [
+    {
+      "id": "sub_result",
+      "label": "NLI Result",
+      "role": "subject",
+      "asset_type": "text",
+      "inputs": [{"id": "in", "label": "Text", "type": "text"}],
+      "outputs": [{"id": "out", "label": "Text", "type": "text"}],
+      "x": 840, "y": 200
+    }
+  ],
+  "edges": [
+    {
+      "id": "e1",
+      "from_node_id": "ref_premise", "from_port_id": "out",
+      "to_node_id": "op_classify", "to_port_id": "premise",
+      "type": "text"
+    },
+    {
+      "id": "e2",
+      "from_node_id": "ref_hypothesis", "from_port_id": "out",
+      "to_node_id": "op_classify", "to_port_id": "hypothesis",
+      "type": "text"
+    },
+    {
+      "id": "e3",
+      "from_node_id": "op_classify", "from_port_id": "out_0",
+      "to_node_id": "sub_result", "to_port_id": "in",
+      "type": "text"
+    }
+  ],
+  "view": {"default": "canvas"}
+}