diff --git a/.github/workflows/benchmark_models.yml b/.github/workflows/benchmark_models.yml
index 2a5e9283..f68b86df 100644
--- a/.github/workflows/benchmark_models.yml
+++ b/.github/workflows/benchmark_models.yml
@@ -64,11 +64,14 @@ jobs:
         BM_REPO_URL: https://github.com/Benchmarking-Initiative/Benchmark-Models-PEtab.git
       run: |
         cd $PARPE_BASE/benchmark_collection \
-          && git clone --depth 1 $BM_REPO_URL
+          && git clone --depth 1 $BM_REPO_URL \
+          && echo "BENCHMARK_COLLECTION=$(pwd)/Benchmark-Models-PEtab/Benchmark-Models/" >> $GITHUB_ENV
 
     - name: Benchmark models --- tests
       run: |
         cd $PARPE_BASE/benchmark_collection \
-          && AMICI_PARALLEL_COMPILE="" \
-             BENCHMARK_COLLECTION="$(pwd)/Benchmark-Models-PEtab/Benchmark-Models/" \
-             $PARPE_BASE/misc/run_in_venv.sh $PARPE_BASE/build/venv ./all.sh
+          && . $PARPE_BASE/build/venv/bin/activate \
+          && python -m pip uninstall -y petab \
+          && python -m pip install https://github.com/PEtab-dev/libpetab-python/archive/develop.zip \
+          && python priors_to_measurements.py \
+          && AMICI_PARALLEL_COMPILE="" ./all.sh
diff --git a/benchmark_collection/priors_to_measurements.py b/benchmark_collection/priors_to_measurements.py
new file mode 100755
index 00000000..dfef202a
--- /dev/null
+++ b/benchmark_collection/priors_to_measurements.py
@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+"""For all benchmark collection problems, convert objectivePriors to
+measurements."""
+import os
+import sys
+from pathlib import Path
+from petab.v1 import (Problem, write_observable_df, write_measurement_df,
+                      write_parameter_df)
+from petab.v1.priors import priors_to_measurements
+from petab.v1.yaml import load_yaml
+from petab.v1.C import (MEASUREMENT_FILES, OBSERVABLE_FILES, PARAMETER_FILE,
+                        PROBLEMS)
+
+
+def get_problems_dir() -> Path:
+    """Get the directory containing the benchmark collection problems."""
+    if len(sys.argv) == 2:
+        problems_dir = sys.argv[1]
+    elif "BENCHMARK_COLLECTION" in os.environ:
+        problems_dir = os.environ["BENCHMARK_COLLECTION"]
+    else:
+        print("Usage: prior_to_measurements.py <benchmark_collection_dir>")
+        sys.exit(1)
+
+    problems_dir = Path(problems_dir)
+    if not problems_dir.is_dir():
+        print(f"Directory {problems_dir} does not exist.")
+        sys.exit(1)
+
+    return problems_dir
+
+
+def save_problem(yaml_path: Path, petab_problem: Problem):
+    """Save the updated PEtab problem."""
+    # only meausrements, observables, and parameters are changed
+    config = load_yaml(yaml_path)
+    assert len(config[PROBLEMS]) == 1
+    problem_config = config[PROBLEMS][0]
+    assert len(problem_config[MEASUREMENT_FILES]) == 1
+    assert len(problem_config[OBSERVABLE_FILES]) == 1
+    parameter_file = yaml_path.parent / config[PARAMETER_FILE]
+    measurement_file = yaml_path.parent / problem_config[MEASUREMENT_FILES][0]
+    observable_file = yaml_path.parent / problem_config[OBSERVABLE_FILES][0]
+    # back up the original files
+    parameter_file.rename(parameter_file.with_suffix(".tsv.bak"))
+    measurement_file.rename(measurement_file.with_suffix(".tsv.bak"))
+    observable_file.rename(observable_file.with_suffix(".tsv.bak"))
+    # write the updated files
+    write_parameter_df(petab_problem.parameter_df, parameter_file)
+    write_observable_df(petab_problem.observable_df, observable_file)
+    write_measurement_df(petab_problem.measurement_df, measurement_file)
+
+
+def main():
+    problems_dir = get_problems_dir()
+
+    for problem_dir in sorted(problems_dir.iterdir()):
+        if not problem_dir.is_dir():
+            continue
+
+        problem_id = problem_dir.name
+        yaml_path = problem_dir / f"{problem_id}.yaml"
+
+        if not yaml_path.is_file():
+            continue
+        print(f"{problem_id}...")
+        petab_problem = Problem.from_yaml(yaml_path)
+        num_measurements_old = len(petab_problem.measurement_df)
+
+        try:
+            petab_problem = priors_to_measurements(petab_problem)
+            num_measurements_new = len(petab_problem.measurement_df)
+            if num_measurements_new != num_measurements_old:
+                print(f"\tConverting {problem_id}...")
+                save_problem(yaml_path, petab_problem)
+            else:
+                print(f"\tNothing to do for {problem_id}.")
+        except NotImplementedError as e:
+            print(f"\t{problem_id}: {e}")
+
+
+if __name__ == "__main__":
+    main()