Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -849,8 +849,8 @@ def check_model_availability(models, token):
|
|
849 |
return availability_results
|
850 |
|
851 |
# Streamlit UI
|
852 |
-
st.title("AutoBench 1.0")
|
853 |
-
|
854 |
|
855 |
# Setup sidebar for configuration
|
856 |
st.sidebar.header("Configuration")
|
@@ -948,7 +948,7 @@ if st.sidebar.button("Start Benchmark"):
|
|
948 |
# Run the benchmark
|
949 |
try:
|
950 |
# Update status
|
951 |
-
status_text.text("Benchmark running...\n\nThis will take a few minutes depending on the number of models and iterations chosen. If you want to follow progress in the question-answer-ranking process per each iteration, check the container log (above, next to the \"running\" button")
|
952 |
|
953 |
# Run benchmark and get results
|
954 |
results, cumulative_avg_rank, total_successful = run_benchmark(
|
|
|
849 |
return availability_results
|
850 |
|
851 |
# Streamlit UI
|
852 |
+
st.title("AutoBench 1.0 Lite")
|
853 |
+
status_text.text("A Many-Model-As-Judge system to generate a customizable LLM benchmark.\nChose the models you want to evaluate (at least 2) and they will rank each other against the selected topics. But first check if models are available (this will depend on your Hugging face account. Premium is strongly recomended)")
|
854 |
|
855 |
# Setup sidebar for configuration
|
856 |
st.sidebar.header("Configuration")
|
|
|
948 |
# Run the benchmark
|
949 |
try:
|
950 |
# Update status
|
951 |
+
status_text.text("Benchmark running...\n\nThis will take a few minutes depending on the number of models and iterations chosen. If you want to follow progress in the question-answer-ranking process per each iteration, check the container log (above, next to the \"running\" button)")
|
952 |
|
953 |
# Run benchmark and get results
|
954 |
results, cumulative_avg_rank, total_successful = run_benchmark(
|