change mlx-community/Meta-Llama-3.1-70B-Instruct-bf16-CORRECTED to mlx-community/Meta-Llama-3.1-70B-Instruct-bf16 since it works now

8d524bfe710e4e8916a5c08e7b893caedb1155d8 · 2024-11-03 18:29:01 -0500 · Alex Cheema

Files touched

M exo/models.py

Diff

commit 8d524bfe710e4e8916a5c08e7b893caedb1155d8
Author: Alex Cheema <alexcheema123@gmail.com>
Date:   Sun Nov 3 18:29:01 2024 -0500

    change mlx-community/Meta-Llama-3.1-70B-Instruct-bf16-CORRECTED to mlx-community/Meta-Llama-3.1-70B-Instruct-bf16 since it works now
---
 exo/models.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/exo/models.py b/exo/models.py
index 5791bdd5..dce0693c 100644
--- a/exo/models.py
+++ b/exo/models.py
@@ -13,7 +13,7 @@ model_base_shards = {
     "TinygradDynamicShardInferenceEngine": Shard(model_id="NousResearch/Meta-Llama-3.1-70B-Instruct", start_layer=0, end_layer=0, n_layers=80),
   },
   "llama-3.1-70b-bf16": {
-    "MLXDynamicShardInferenceEngine": Shard(model_id="mlx-community/Meta-Llama-3.1-70B-Instruct-bf16-CORRECTED", start_layer=0, end_layer=0, n_layers=80),
+    "MLXDynamicShardInferenceEngine": Shard(model_id="mlx-community/Meta-Llama-3.1-70B-Instruct-bf16", start_layer=0, end_layer=0, n_layers=80),
     "TinygradDynamicShardInferenceEngine": Shard(model_id="NousResearch/Meta-Llama-3.1-70B-Instruct", start_layer=0, end_layer=0, n_layers=80),
   },
   "llama-3.1-405b": {"MLXDynamicShardInferenceEngine": Shard(model_id="mlx-community/Meta-Llama-3.1-405B-4bit", start_layer=0, end_layer=0, n_layers=126),},

← ab91f202 Revert "add max-caches option" · back to Exo · revert back to CORRECTED 436709e5 →