diff --git a/.github/workflows/benchmark.yml b/.github/workflows/benchmark.yml
index 5dbc500b..776bdee6 100644
--- a/.github/workflows/benchmark.yml
+++ b/.github/workflows/benchmark.yml
@@ -71,11 +71,6 @@ jobs:
         python -m pip install maturin
         rustup target add x86_64-unknown-linux-musl
 
-    # we need this for brun and opc
-    - name: install clvm_tools
-      run: |
-        cargo install clvm_tools_rs
-
     - name: Build
       env:
         CC: gcc
diff --git a/benches/run-program.rs b/benches/run-program.rs
index ef1cb729..94b9da97 100644
--- a/benches/run-program.rs
+++ b/benches/run-program.rs
@@ -224,14 +224,13 @@ fn run_program_benchmark(c: &mut Criterion) {
     ] {
         a.restore_checkpoint(&test_case_checkpoint);
 
-        println!("benchmark/{test}.hex");
-        let prg = read_to_string(format!("benchmark/{test}.hex"))
-            .expect("failed to load benchmark program");
-        let prg = hex::decode(prg.trim()).expect("invalid hex in benchmark program");
-        let prg = node_from_bytes(&mut a, &prg[..]).expect("failed to parse benchmark program");
-        let env = make_env(&mut a);
-        let iter_checkpoint = a.checkpoint();
         group.bench_function(*test, |b| {
+            let prg = read_to_string(format!("benchmark/{test}.hex"))
+                .expect("failed to load benchmark program");
+            let prg = hex::decode(prg.trim()).expect("invalid hex in benchmark program");
+            let prg = node_from_bytes(&mut a, &prg[..]).expect("failed to parse benchmark program");
+            let env = make_env(&mut a);
+            let iter_checkpoint = a.checkpoint();
             b.iter(|| {
                 a.restore_checkpoint(&iter_checkpoint);
                 let start = Instant::now();
diff --git a/tests/run-programs.py b/tests/run-programs.py
index dd85b3fc..aebb0b13 100755
--- a/tests/run-programs.py
+++ b/tests/run-programs.py
@@ -7,6 +7,7 @@
 import platform
 from colorama import init, Fore, Style
 from run import run_clvm
+from os.path import isfile
 
 init()
 ret = 0
@@ -14,22 +15,27 @@
 for fn in glob.glob('programs/large-atom-*.hex.invalid'):
 
     try:
+        print(fn)
         run_clvm(fn)
         ret = 1
-        print("FAILED: expected parse failure")
+        print(Fore.RED + "FAILED: expected parse failure" + Style.RESET_ALL)
     except Exception as e:
-        print("expected failure: %s" % e)
+        print(Fore.GREEN + f"OK: expected: {e}" + Style.RESET_ALL)
 
 
 for fn in glob.glob('programs/*.clvm'):
 
     hexname = fn[:-4] + 'hex'
+    if isfile(hexname):
+        continue
     with open(hexname, 'w+') as out:
         proc = subprocess.Popen(['opc', fn], stdout=out)
         proc.wait()
 
 for fn in glob.glob('programs/*.env'):
     hexenv = fn + 'hex'
+    if isfile(hexenv):
+        continue
     with open(hexenv, 'w+') as out:
         proc = subprocess.Popen(['opc', fn], stdout=out)
         proc.wait()
@@ -38,7 +44,6 @@
 
     hexenv = hexname[:-3] + 'envhex'
 
-#    command = ['brun', '-m', '11000000000', '-c', '--backend=rust', '--quiet', '--time', '--hex', hexname, hexenv]
     command = ['./run.py', hexname, hexenv]
 
     # prepend the size command, to measure RSS