add qwen-2.5-1.5b, qwen-2.5-3b, qwen-2.5-32b

c609c05e408308a5009ec4417f119af3aab48ffc · 2024-12-24 01:50:12 +0000 · Alex Cheema

Files touched

M exo/models.py

Diff

commit c609c05e408308a5009ec4417f119af3aab48ffc
Author: Alex Cheema <alexcheema123@gmail.com>
Date:   Tue Dec 24 01:50:12 2024 +0000

    add qwen-2.5-1.5b, qwen-2.5-3b, qwen-2.5-32b
---
 exo/models.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/exo/models.py b/exo/models.py
index 0f984d48..226046fb 100644
--- a/exo/models.py
+++ b/exo/models.py
@@ -92,14 +92,17 @@ model_cards = {
   "llava-1.5-7b-hf": { "layers": 32, "repo": { "MLXDynamicShardInferenceEngine": "llava-hf/llava-1.5-7b-hf", }, },
   ### qwen
   "qwen-2.5-0.5b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-0.5B-Instruct-4bit", }, },
+  "qwen-2.5-1.5b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-1.5B-Instruct-4bit", }, },
   "qwen-2.5-coder-1.5b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-1.5B-Instruct-4bit", }, },
+  "qwen-2.5-3b": { "layers": 36, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-3B-Instruct-4bit", }, },
   "qwen-2.5-coder-3b": { "layers": 36, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-3B-Instruct-4bit", }, },
-  "qwen-2.5-coder-7b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-7B-Instruct-4bit", }, },
-  "qwen-2.5-coder-14b": { "layers": 48, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-14B-Instruct-4bit", }, },
-  "qwen-2.5-coder-32b": { "layers": 64, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-32B-Instruct-4bit", }, },
   "qwen-2.5-7b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-7B-Instruct-4bit", }, },
+  "qwen-2.5-coder-7b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-7B-Instruct-4bit", }, },
   "qwen-2.5-math-7b": { "layers": 28, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Math-7B-Instruct-4bit", }, },
   "qwen-2.5-14b": { "layers": 48, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-14B-Instruct-4bit", }, },
+  "qwen-2.5-coder-14b": { "layers": 48, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-14B-Instruct-4bit", }, },
+  "qwen-2.5-32b": { "layers": 64, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-32B-Instruct-4bit", }, },
+  "qwen-2.5-coder-32b": { "layers": 64, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Coder-32B-Instruct-4bit", }, },
   "qwen-2.5-72b": { "layers": 80, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-72B-Instruct-4bit", }, },
   "qwen-2.5-math-72b": { "layers": 80, "repo": { "MLXDynamicShardInferenceEngine": "mlx-community/Qwen2.5-Math-72B-Instruct-4bit", }, },
   ### nemotron

← cde912de - Use `#!/usr/bin/env bash` instead of `#!/bin/bash` for bet · back to Exo · add chatgpt-api-compatible tools for function calling 456fbdd2 →