Add files using upload-large-folder tool

README.md

@@ -42,13 +42,11 @@ TRIPLES:
 Output as markdown table with Type and Triple columns.
 ```
 
-## ms-swift PtEngine
+## ms-swift
 
 ```python
 import torch
-from huggingface_hub import snapshot_download
-from swift.llm import InferRequest, PtEngine, RequestConfig, get_model_tokenizer, get_template
-from swift.tuners import Swift
+from swift.infer_engine import InferRequest, RequestConfig, TransformersEngine
 
 BASE_MODEL = "google/gemma-3-12b-it"
 ADAPTER_ID = "Loria-MosAIk/xqdt-e2e-gemma3-12b"
@@ -59,18 +57,20 @@ TRIPLES:
 1. [S] Blue Spice [P] area [O] city centre
 2. [S] Blue Spice [P] eat type [O] coffee shop
 Output as markdown table with Type and Triple columns."""
+MESSAGES = [{"role": "user", "content": QUERY}]
 
-adapter_path = snapshot_download(ADAPTER_ID)
-model, tokenizer = get_model_tokenizer(
+engine = TransformersEngine(
     BASE_MODEL,
-    model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16},
+    adapters=[ADAPTER_ID],
+    max_batch_size=1,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    template_type="gemma3_text",
+    use_hf=True,
 )
-model = Swift.from_pretrained(model, model_id=adapter_path, adapter_name="default")
-template = get_template("gemma3_text", tokenizer, default_system=None)
-engine = PtEngine.from_model_template(model, template, max_batch_size=1)
 response = engine.infer(
-    [InferRequest(messages=[{"role": "user", "content": QUERY}])],
-    RequestConfig(max_tokens=1024, temperature=0.3),
+    [InferRequest(messages=MESSAGES)],
+    RequestConfig(max_tokens=1024, temperature=0.3, seed=2023),
     use_tqdm=False,
 )[0]
 print(response.choices[0].message.content)