mega vicuna merge pt 2 (#1685)

.github/workflows/test-models.yml

@@ -115,6 +115,7 @@ jobs:
         pytest --forked --benchmark=native --ci --ci_sha=${SHORT_SHA} --update_tank --tank_url="gs://shark_tank/nightly/" -k cpu 
         gsutil cp ./bench_results.csv gs://shark-public/builder/bench_results/${DATE}/bench_results_cpu_${SHORT_SHA}.csv
         gsutil cp gs://shark-public/builder/bench_results/${DATE}/bench_results_cpu_${SHORT_SHA}.csv gs://shark-public/builder/bench_results/latest/bench_results_cpu_latest.csv
+        python build_tools/vicuna_testing.py
 
     - name: Validate Models on NVIDIA GPU
       if: matrix.suite == 'cuda'

build_tools/vicuna_testing.py

@@ -0,0 +1,14 @@
+import os
+from sys import executable
+import subprocess
+from apps.language_models.scripts import vicuna
+
+
+def test_loop():
+    precisions = ["fp16", "int8", "int4"]
+    devices = ["cpu"]
+    for precision in precisions:
+        for device in devices:
+            model = vicuna.UnshardedVicuna(device=device, precision=precision)
+            model.compile()
+            del model