Remove model size restriction

app.py

@@ -120,16 +120,6 @@ leaderboard_df = original_df.copy()
 ) = get_evaluation_queue_df(eval_queue, eval_queue_private, EVAL_REQUESTS_PATH, EVAL_COLS)
 
 ## INTERACTION FUNCTIONS
-def is_model_acceptable(model: str) -> bool:
-    # regex for something like '13b'
-    pattern = r'(\d+)[bB]'
-    values = re.findall(pattern, model)
-    for val in values:
-        if int(val) <= 13:
-            return True
-    
-    return values == []
-
 def add_new_eval(
     model: str,
     base_model: str,
@@ -139,9 +129,6 @@ def add_new_eval(
     weight_type: str,
     model_type: str,
 ):
-    if not is_model_acceptable(model):
-        return styled_error("Please submit a model with less than 13B parameters.")
-
     precision = precision.split(" ")[0]
     current_time = datetime.now(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ")
 
@@ -261,8 +248,8 @@ NUMERIC_INTERVALS = {
     "~3B": pd.Interval(1.5, 5, closed="right"),
     "~7B": pd.Interval(6, 11, closed="right"),
     "~13B": pd.Interval(12, 15, closed="right"),
-    # "~35B": pd.Interval(16, 55, closed="right"),
-    # "60B+": pd.Interval(55, 10000, closed="right"),
+    "~35B": pd.Interval(16, 55, closed="right"),
+    "60B+": pd.Interval(55, 10000, closed="right"),
 }
 
 def filter_models(
@@ -571,7 +558,6 @@ with demo:
                     base_model_name_textbox = gr.Textbox(label="Base model (for delta or adapter weights)")
 
             submit_button = gr.Button("Submit Evalulation!")
-            gr.Markdown("_We accept models with less than or equal to 13B parameters now._")
             submission_result = gr.Markdown()
             submit_button.click(
                 add_new_eval,

src/assets/text_content.py

@@ -7,7 +7,6 @@ INTRODUCTION_TEXT = f"""
 🚀 The Open Ko-LLM Leaderboard 🇰🇷 objectively evaluates the performance of Korean Large Language Model (LLM).
 
 When you submit a model on the "Submit here!" page, it is automatically evaluated. The GPU used for evaluation is operated with the support of KT.
-We accept models with less than or equal to 13B parameters now.
 The data used for evaluation consists of datasets to assess reasoning, language understanding, hallucination, and commonsense. 
 The evaluation dataset is exclusively private and only available for evaluation process.
 More detailed information about the benchmark dataset is provided on the “About” page.