diff --git a/bench/eval_reme_old.py b/bench_old/eval_reme_old.py
similarity index 100%
rename from bench/eval_reme_old.py
rename to bench_old/eval_reme_old.py
diff --git a/bench/halumem/analyze_dataset_stats.py b/bench_old/halumem/analyze_dataset_stats.py
similarity index 100%
rename from bench/halumem/analyze_dataset_stats.py
rename to bench_old/halumem/analyze_dataset_stats.py
diff --git a/bench/halumem/analyze_results.py b/bench_old/halumem/analyze_results.py
similarity index 100%
rename from bench/halumem/analyze_results.py
rename to bench_old/halumem/analyze_results.py
diff --git a/bench/halumem/compute_qa_stats_v4.py b/bench_old/halumem/compute_qa_stats_v4.py
similarity index 100%
rename from bench/halumem/compute_qa_stats_v4.py
rename to bench_old/halumem/compute_qa_stats_v4.py
diff --git a/bench/halumem/compute_stats_from_tmp.py b/bench_old/halumem/compute_stats_from_tmp.py
similarity index 100%
rename from bench/halumem/compute_stats_from_tmp.py
rename to bench_old/halumem/compute_stats_from_tmp.py
diff --git a/bench/halumem/eval_baseline_simple.py b/bench_old/halumem/eval_baseline_simple.py
similarity index 100%
rename from bench/halumem/eval_baseline_simple.py
rename to bench_old/halumem/eval_baseline_simple.py
diff --git a/bench/halumem/eval_reme.py b/bench_old/halumem/eval_reme.py
similarity index 100%
rename from bench/halumem/eval_reme.py
rename to bench_old/halumem/eval_reme.py
diff --git a/bench/halumem/eval_reme_simple.py b/bench_old/halumem/eval_reme_simple.py
similarity index 100%
rename from bench/halumem/eval_reme_simple.py
rename to bench_old/halumem/eval_reme_simple.py
diff --git a/bench/halumem/eval_reme_simple_v3.py b/bench_old/halumem/eval_reme_simple_v3.py
similarity index 100%
rename from bench/halumem/eval_reme_simple_v3.py
rename to bench_old/halumem/eval_reme_simple_v3.py
diff --git a/bench/halumem/eval_reme_simple_v4.py b/bench_old/halumem/eval_reme_simple_v4.py
similarity index 100%
rename from bench/halumem/eval_reme_simple_v4.py
rename to bench_old/halumem/eval_reme_simple_v4.py
diff --git a/bench/halumem/eval_tools.py b/bench_old/halumem/eval_tools.py
similarity index 100%
rename from bench/halumem/eval_tools.py
rename to bench_old/halumem/eval_tools.py
diff --git a/bench/halumem/halumem.yaml b/bench_old/halumem/halumem.yaml
similarity index 100%
rename from bench/halumem/halumem.yaml
rename to bench_old/halumem/halumem.yaml
diff --git a/bench/halumem/llms.py b/bench_old/halumem/llms.py
similarity index 100%
rename from bench/halumem/llms.py
rename to bench_old/halumem/llms.py
diff --git a/bench/human_in_the_loop/__init__.py b/bench_old/human_in_the_loop/__init__.py
similarity index 100%
rename from bench/human_in_the_loop/__init__.py
rename to bench_old/human_in_the_loop/__init__.py
diff --git a/bench/human_in_the_loop/compute_qa_stats.py b/bench_old/human_in_the_loop/compute_qa_stats.py
similarity index 100%
rename from bench/human_in_the_loop/compute_qa_stats.py
rename to bench_old/human_in_the_loop/compute_qa_stats.py
diff --git a/bench/human_in_the_loop/eval.yaml b/bench_old/human_in_the_loop/eval.yaml
similarity index 100%
rename from bench/human_in_the_loop/eval.yaml
rename to bench_old/human_in_the_loop/eval.yaml
diff --git a/bench/human_in_the_loop/reevaluate_qa.py b/bench_old/human_in_the_loop/reevaluate_qa.py
similarity index 100%
rename from bench/human_in_the_loop/reevaluate_qa.py
rename to bench_old/human_in_the_loop/reevaluate_qa.py
diff --git a/bench/human_in_the_loop2/__init__.py b/bench_old/human_in_the_loop2/__init__.py
similarity index 100%
rename from bench/human_in_the_loop2/__init__.py
rename to bench_old/human_in_the_loop2/__init__.py
diff --git a/bench/human_in_the_loop2/compute_qa_stats.py b/bench_old/human_in_the_loop2/compute_qa_stats.py
similarity index 100%
rename from bench/human_in_the_loop2/compute_qa_stats.py
rename to bench_old/human_in_the_loop2/compute_qa_stats.py
diff --git a/bench/human_in_the_loop2/eval.yaml b/bench_old/human_in_the_loop2/eval.yaml
similarity index 100%
rename from bench/human_in_the_loop2/eval.yaml
rename to bench_old/human_in_the_loop2/eval.yaml
diff --git a/bench/human_in_the_loop2/reevaluate_qa.py b/bench_old/human_in_the_loop2/reevaluate_qa.py
similarity index 100%
rename from bench/human_in_the_loop2/reevaluate_qa.py
rename to bench_old/human_in_the_loop2/reevaluate_qa.py