tests

45142dab263cd6bbcab731d10867b83c3cb2f547 · 2024-08-06 12:47:21 +0100 · Alex Cheema

Files touched

M exo/inference/test_inference_engine.py

Diff

commit 45142dab263cd6bbcab731d10867b83c3cb2f547
Author: Alex Cheema <alexcheema123@gmail.com>
Date:   Tue Aug 6 12:47:21 2024 +0100

    tests
---
 exo/inference/test_inference_engine.py | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/exo/inference/test_inference_engine.py b/exo/inference/test_inference_engine.py
index 76cf44c0..58eecc8a 100644
--- a/exo/inference/test_inference_engine.py
+++ b/exo/inference/test_inference_engine.py
@@ -41,17 +41,17 @@ async def test_inference_engine(inference_engine_1: InferenceEngine, inference_e
   assert np.array_equal(next_resp_full, resp4)
 
 
-# asyncio.run(
-#   test_inference_engine(
-#     MLXDynamicShardInferenceEngine(),
-#     MLXDynamicShardInferenceEngine(),
-#     "mlx-community/Meta-Llama-3-8B-Instruct-4bit",
-#   )
-# )
+asyncio.run(
+  test_inference_engine(
+    MLXDynamicShardInferenceEngine(),
+    MLXDynamicShardInferenceEngine(),
+    "mlx-community/Meta-Llama-3-8B-Instruct-4bit",
+  )
+)
 
 # TODO: Need more memory or a smaller model
-asyncio.run(test_inference_engine(
-    TinygradDynamicShardInferenceEngine(),
-    TinygradDynamicShardInferenceEngine(),
-    "llama3-8b-sfr",
-))
+# asyncio.run(test_inference_engine(
+#     TinygradDynamicShardInferenceEngine(),
+#     TinygradDynamicShardInferenceEngine(),
+#     "mlx-community/Meta-Llama-3-8B-Instruct",
+# ))

← 545a486e separate hf_helpers, make extra dir with download_hf script, · back to Exo · bring tinygrad to parity with mlx on llama models, show prog d22ed12e →