fix(model): modify model name kimi2deepseek

src/parallax/models/kimi_k2.py → src/parallax/models/deepseek_v3.py

@@ -11,7 +11,7 @@
 from mlx_lm.models.deepseek_v3 import ModelArgs
 
 
-class ParallaxKimiK2Attention(MLXDeepseekV3Attention):
+class ParallaxDeepSeekV3Attention(MLXDeepseekV3Attention):
     """A custom attention module for Parallax, extending the DeepseekV3 Attention class.
 
     We apply explicit KV cache handling and passing in `offset` directly from Request.
@@ -95,14 +95,14 @@ def __call__(
         return self.o_proj(output), (mx.concatenate([k_nope, k_pe], axis=-1), values)
 
 
-class ParallaxKimiK2Block(MLXDeepseekV3Block):
+class ParallaxDeepSeekV3Block(MLXDeepseekV3Block):
     """A custom transformer block for Parallax, extending the Qwen3 Block class.
     This version handles the KV cache explicitly and returns new K and V states.
     """
 
     def __init__(self, args: ModelArgs, layer_idx: int):
         super().__init__(args, layer_idx=layer_idx)
-        self.self_attn = ParallaxKimiK2Attention(args)
+        self.self_attn = ParallaxDeepSeekV3Attention(args)
 
     def __call__(
         self,
@@ -124,4 +124,4 @@ def get_architecture(cls):
         return "DeepseekV3ForCausalLM"
 
 
-EntryClass = ParallaxKimiK2Block
+EntryClass = ParallaxDeepSeekV3Block

src/parallax/server/executor.py

@@ -143,6 +143,7 @@ def __init__(
         self.finished_batch = []
         self.start_layer = start_layer
         self.end_layer = end_layer
+
         self.is_first_peer = start_layer == 0
         self.is_last_peer = end_layer == self.config.get("num_hidden_layers")
         self.num_shard_layers = end_layer - start_layer