Update app.py

app.py

@@ -20,21 +20,33 @@ def generate(
     prompt = pipe.tokenizer.apply_chat_template(message_template, tokenize=False, add_generation_prompt=True)
     
     if model_name == "Felladrin/Pythia-31M-Chat-v1":
-        outputs = pipe(prompt, max_new_tokens=250, use_cache=True, penalty_alpha=0.5, top_k=2, repetition_penalty=1.0016)
+        outputs = pipe(prompt, max_new_tokens=512, use_cache=True, penalty_alpha=0.5, top_k=2, repetition_penalty=1.0016)
     elif model_name == "Felladrin/Llama-68M-Chat-v1":
-        outputs = pipe(prompt, max_new_tokens=250, use_cache=True, penalty_alpha=0.5, top_k=4, repetition_penalty=1.043)
+        outputs = pipe(prompt, max_new_tokens=512, use_cache=True, penalty_alpha=0.5, top_k=4, repetition_penalty=1.043)
     elif model_name == "Felladrin/Smol-Llama-101M-Chat-v1":
-        outputs = pipe(prompt, max_new_tokens=250, use_cache=True, penalty_alpha=0.5, top_k=4, repetition_penalty=1.105)
+        outputs = pipe(prompt, max_new_tokens=512, use_cache=True, penalty_alpha=0.5, top_k=4, repetition_penalty=1.105)
     elif model_name == "Felladrin/Llama-160M-Chat-v1":
-        outputs = pipe(prompt, max_new_tokens=250, use_cache=True, penalty_alpha=0.5, top_k=4, repetition_penalty=1.01)
+        outputs = pipe(prompt, max_new_tokens=512, use_cache=True, penalty_alpha=0.5, top_k=4, repetition_penalty=1.01)
     elif model_name == "Felladrin/TinyMistral-248M-Chat-v4":
-        outputs = pipe(prompt, max_new_tokens=250, use_cache=True, penalty_alpha=0.5, top_k=5, repetition_penalty=1.0)
+        outputs = pipe(prompt, max_new_tokens=512, use_cache=True, penalty_alpha=0.5, top_k=5, repetition_penalty=1.0)
+    elif "Felladrin/Minueza-2-96M-Instruct-Variant" in model_name:
+        outputs = pipe(prompt, max_new_tokens=512, use_cache=True, do_sample=True, temperature=0.7, top_p=0.9, top_k=0, min_p=0.1, repetition_penalty=1.17)
     else:
-        outputs = pipe(prompt, max_new_tokens=250, do_sample=True, temperature=0.65, top_k=35, top_p=0.55, repetition_penalty=1.176)
+        outputs = pipe(prompt, max_new_tokens=512, do_sample=True, temperature=0.65, top_k=35, top_p=0.55, repetition_penalty=1.176)
         
     return outputs[0]["generated_text"]
 
 model_choices = [
+    "Felladrin/Minueza-2-96M-Instruct-Variant-10",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-09",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-08",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-07",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-06",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-05",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-04",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-03",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-02",
+    "Felladrin/Minueza-2-96M-Instruct-Variant-01",
     "Felladrin/Llama-160M-Chat-v1",
     "Felladrin/TinyMistral-248M-Chat-v4",
     "Felladrin/Llama-68M-Chat-v1",