vllm.model_executor.models.mellum

Classes:

• MellumAttention –

  Differences from Qwen3MoeAttention:

• MellumDecoderLayer –

  Differences from Qwen3MoeDecoderLayer:

• MellumForCausalLM –

  Differences from Qwen3MoeForCausalLM:

• MellumModel –

  Differences from Qwen3MoeModel:

MellumAttention

Bases: Qwen3MoeAttention

Differences from Qwen3MoeAttention: - Supports per_layer_sliding_window for Attention.

Source code in vllm/model_executor/models/mellum.py

class MellumAttention(Qwen3MoeAttention):
    """Differences from `Qwen3MoeAttention`:
    - Supports `per_layer_sliding_window` for `Attention`.
    """

    def __init__(
        self,
        hidden_size: int,
        num_heads: int,
        num_kv_heads: int,
        rope_parameters: dict[str, Any],
        max_position_embeddings: int = 8192,
        head_dim: int | None = None,
        rms_norm_eps: float = 1e-06,
        qkv_bias: bool = False,
        cache_config: Any | None = None,
        quant_config: Any | None = None,
        prefix: str = "",
        dual_chunk_attention_config: dict[str, Any] | None = None,
        per_layer_sliding_window: int | None = None,
    ) -> None:
        nn.Module.__init__(self)

        self.hidden_size = hidden_size
        tp_size = get_tensor_model_parallel_world_size()
        self.total_num_heads = num_heads
        assert self.total_num_heads % tp_size == 0
        self.num_heads = self.total_num_heads // tp_size
        self.total_num_kv_heads = num_kv_heads
        if self.total_num_kv_heads >= tp_size:
            assert self.total_num_kv_heads % tp_size == 0
        else:
            assert tp_size % self.total_num_kv_heads == 0
        self.num_kv_heads = max(1, self.total_num_kv_heads // tp_size)
        self.head_dim = head_dim or (hidden_size // self.total_num_heads)
        self.q_size = self.num_heads * self.head_dim
        self.kv_size = self.num_kv_heads * self.head_dim
        self.scaling = self.head_dim**-0.5
        self.max_position_embeddings = max_position_embeddings
        self.dual_chunk_attention_config = dual_chunk_attention_config

        self.qkv_proj = QKVParallelLinear(
            hidden_size,
            self.head_dim,
            self.total_num_heads,
            self.total_num_kv_heads,
            bias=qkv_bias,
            quant_config=quant_config,
            prefix=f"{prefix}.qkv_proj",
        )

        self.o_proj = RowParallelLinear(
            self.total_num_heads * self.head_dim,
            hidden_size,
            bias=False,
            quant_config=quant_config,
            prefix=f"{prefix}.o_proj",
        )

        self.rotary_emb = get_rope(
            self.head_dim,
            max_position=max_position_embeddings,
            rope_parameters=rope_parameters,
            dual_chunk_attention_config=dual_chunk_attention_config,
        )
        if dual_chunk_attention_config:
            self.attn = Attention(
                self.num_heads,
                self.head_dim,
                self.scaling,
                num_kv_heads=self.num_kv_heads,
                cache_config=cache_config,
                quant_config=quant_config,
                per_layer_sliding_window=per_layer_sliding_window,
                prefix=f"{prefix}.attn",
                layer_idx=extract_layer_index(prefix),
                dual_chunk_attention_config=dual_chunk_attention_config,
            )
        else:
            self.attn = Attention(
                self.num_heads,
                self.head_dim,
                self.scaling,
                num_kv_heads=self.num_kv_heads,
                cache_config=cache_config,
                quant_config=quant_config,
                per_layer_sliding_window=per_layer_sliding_window,
                prefix=f"{prefix}.attn",
            )

        self.q_norm = RMSNorm(self.head_dim, eps=rms_norm_eps)
        self.k_norm = RMSNorm(self.head_dim, eps=rms_norm_eps)

MellumDecoderLayer

Bases: Qwen3MoeDecoderLayer

Differences from Qwen3MoeDecoderLayer: - Supports interleaved SWA and per-layer RoPE scaling.

Source code in vllm/model_executor/models/mellum.py

class MellumDecoderLayer(Qwen3MoeDecoderLayer):
    """Differences from `Qwen3MoeDecoderLayer`:
    - Supports interleaved SWA and per-layer RoPE scaling.
    """

    def __init__(self, vllm_config: VllmConfig, prefix: str = "") -> None:
        nn.Module.__init__(self)

        config = vllm_config.model_config.hf_text_config
        cache_config = vllm_config.cache_config
        quant_config = vllm_config.quant_config

        self.hidden_size = config.hidden_size
        max_position_embeddings = getattr(config, "max_position_embeddings", 8192)
        dual_chunk_attention_config = getattr(
            config, "dual_chunk_attention_config", None
        )

        layer_idx = extract_layer_index(prefix)
        layer_type = config.layer_types[layer_idx]
        if layer_type == "sliding_attention":
            sliding_window = getattr(config, "sliding_window", None)
        else:
            sliding_window = None
        rope_parameters = config.rope_parameters[layer_type]

        self.self_attn = MellumAttention(
            hidden_size=self.hidden_size,
            num_heads=config.num_attention_heads,
            num_kv_heads=config.num_key_value_heads,
            rope_parameters=rope_parameters,
            max_position_embeddings=max_position_embeddings,
            rms_norm_eps=config.rms_norm_eps,
            qkv_bias=getattr(config, "attention_bias", False),
            head_dim=getattr(config, "head_dim", None),
            cache_config=cache_config,
            quant_config=quant_config,
            prefix=f"{prefix}.self_attn",
            dual_chunk_attention_config=dual_chunk_attention_config,
            per_layer_sliding_window=sliding_window,
        )

        if config.mlp_layer_types[layer_idx] == "sparse":
            self.mlp = Qwen3MoeSparseMoeBlock(
                vllm_config=vllm_config, prefix=f"{prefix}.mlp"
            )
        else:
            self.mlp = Qwen3MoeMLP(
                hidden_size=config.hidden_size,
                intermediate_size=config.intermediate_size,
                hidden_act=config.hidden_act,
                quant_config=quant_config,
                prefix=f"{prefix}.mlp",
            )

        self.input_layernorm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
        self.post_attention_layernorm = RMSNorm(
            config.hidden_size, eps=config.rms_norm_eps
        )

MellumForCausalLM

Bases: Qwen3MoeForCausalLM

Differences from Qwen3MoeForCausalLM: - Uses MellumModel.

Source code in vllm/model_executor/models/mellum.py

class MellumForCausalLM(Qwen3MoeForCausalLM):
    """Differences from `Qwen3MoeForCausalLM`:
    - Uses `MellumModel`.
    """

    def __init__(self, *, vllm_config: VllmConfig, prefix: str = ""):
        nn.Module.__init__(self)
        config = vllm_config.model_config.hf_text_config
        quant_config = vllm_config.quant_config
        self.config = config
        self.quant_config = quant_config
        if "dense" in getattr(config, "mlp_layer_types", []):
            self.packed_modules_mapping["gate_up_proj"] = ["gate_proj", "up_proj"]
        self.model = MellumModel(
            vllm_config=vllm_config, prefix=maybe_prefix(prefix, "model")
        )
        self.lm_head = ParallelLMHead(
            config.vocab_size,
            config.hidden_size,
            quant_config=quant_config,
            prefix=maybe_prefix(prefix, "lm_head"),
        )
        if self.config.tie_word_embeddings:
            self.lm_head = self.lm_head.tie_weights(self.model.embed_tokens)
        self.logits_processor = LogitsProcessor(config.vocab_size)
        self.make_empty_intermediate_tensors = (
            self.model.make_empty_intermediate_tensors
        )

        self.moe_layers = []
        example_layer = None
        for layer in self.model.layers:
            if isinstance(layer, PPMissingLayer):
                continue

            assert isinstance(layer, Qwen3MoeDecoderLayer)
            if isinstance(layer.mlp, Qwen3MoeSparseMoeBlock):
                example_layer = layer.mlp
                self.moe_layers.append(layer.mlp.experts)

        if example_layer is None:
            raise RuntimeError("No MoE layer found in the model.layers.")

        self.num_moe_layers = len(self.moe_layers)
        self.num_expert_groups = 1
        self.num_shared_experts = 0
        self.num_logical_experts = example_layer.n_logical_experts
        self.num_physical_experts = example_layer.n_physical_experts
        self.num_local_physical_experts = example_layer.n_local_physical_experts
        self.num_routed_experts = example_layer.n_routed_experts
        self.num_redundant_experts = example_layer.n_redundant_experts

MellumModel

Bases: Qwen3MoeModel

Differences from Qwen3MoeModel: - Uses MellumDecoderLayer.

Source code in vllm/model_executor/models/mellum.py

@support_torch_compile
class MellumModel(Qwen3MoeModel):
    """Differences from `Qwen3MoeModel`:
    - Uses `MellumDecoderLayer`.
    """

    def __init__(
        self,
        *,
        vllm_config: VllmConfig,
        prefix: str = "",
    ):
        super().__init__(
            vllm_config=vllm_config,
            prefix=prefix,
            decoder_layer_type=MellumDecoderLayer,
        )