inital test commit

app.py

@@ -0,0 +1,21 @@
+import gradio as gr
+from llama_cpp import Llama
+
+MAX_TOKENS = 64
+
+llm = Llama(model_path="ggml-model-f16-q4_0.bin", n_ctx=2048)
+
+def generate_text(prompt):
+    output = llm(prompt, max_tokens=MAX_TOKENS, echo=False, stop=["</s>", "\n", "User:"])
+    output_text = output['choices'][0]['text']
+    return output_text
+
+description = f"txt2tag 13B base model, max_tokens={MAX_TOKENS}"
+
+gradio_interface = gr.Interface(
+  fn=generate_text,
+  inputs="text",
+  outputs="text",
+  title="txt2tag API",
+)
+gradio_interface.launch()

ggml-model-f16-q4_0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43c1bce0a7cf7c9debec8590c144ef2c831db8616704e698110065072a528a32
+size 7323305088