[Model] Support TP/PP/mamba2 kernel for PLaMo2 (#19674)

Signed-off-by: Shinichi Hemmi <shemmi@preferred.jp>
Signed-off-by: Shinichi Hemmi <50256998+Alnusjaponica@users.noreply.github.com>
Co-authored-by: Calvin Metzger <metzger@preferred.jp>
Co-authored-by: Sixue Wang <cecilwang@preferred.jp>

docs/models/supported_models.md

@@ -389,7 +389,7 @@ th {
 | `PhiMoEForCausalLM` | Phi-3.5-MoE | `microsoft/Phi-3.5-MoE-instruct`, etc. | ✅︎ | ✅︎ | ✅︎ |
 | `Phi4FlashForCausalLM` | Phi-4-mini-flash-reasoning | `microsoft/microsoft/Phi-4-mini-instruct`, etc. | | | |
 | `PersimmonForCausalLM` | Persimmon | `adept/persimmon-8b-base`, `adept/persimmon-8b-chat`, etc. | | ✅︎ | ✅︎ |
-| `Plamo2ForCausalLM` | PLaMo2 | `pfnet/plamo-2-1b`, `pfnet/plamo-2-8b`, etc. | | | |
+| `Plamo2ForCausalLM` | PLaMo2 | `pfnet/plamo-2-1b`, `pfnet/plamo-2-8b`, etc. | | ✅︎ | |
 | `QWenLMHeadModel` | Qwen | `Qwen/Qwen-7B`, `Qwen/Qwen-7B-Chat`, etc. | ✅︎ | ✅︎ | ✅︎ |
 | `Qwen2ForCausalLM` | QwQ, Qwen2 | `Qwen/QwQ-32B-Preview`, `Qwen/Qwen2-7B-Instruct`, `Qwen/Qwen2-7B`, etc. | ✅︎ | ✅︎ | ✅︎ |
 | `Qwen2MoeForCausalLM` | Qwen2MoE | `Qwen/Qwen1.5-MoE-A2.7B`, `Qwen/Qwen1.5-MoE-A2.7B-Chat`, etc. | ✅︎ | ✅︎ | ✅︎ |

tests/distributed/test_pipeline_parallel.py

@@ -175,6 +175,7 @@ TEXT_GENERATION_MODELS = {
     "internlm/internlm2-chat-7b": PPTestSettings.fast(),
     "inceptionai/jais-13b-chat": PPTestSettings.fast(),
     "ai21labs/Jamba-tiny-dev": PPTestSettings.fast(),
+    "pfnet/plamo-2-1b": PPTestSettings.fast(),
     "meta-llama/Llama-3.2-1B-Instruct": PPTestSettings.detailed(),
     # Tests TransformersForCausalLM
     "hmellor/Ilama-3.2-1B": PPTestSettings.fast(),

tests/quantization/test_experts_int8.py

@@ -9,7 +9,7 @@ import pytest
 
 from tests.quantization.utils import is_quant_method_supported
 
-MODELS = ["ai21labs/Jamba-tiny-random"]
+MODELS = ["ai21labs/Jamba-tiny-random", "pfnet/plamo-2-1b"]
 
 
 @pytest.mark.skipif(not is_quant_method_supported("experts_int8"),

vllm/model_executor/models/plamo2.py

@@ -1,7 +1,6 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright contributors to the vLLM project
 """Inference-only PLaMo2 model."""
-import math
 from collections.abc import Iterable
 from typing import Optional
 
@@ -11,30 +10,40 @@ from transformers import PretrainedConfig, PreTrainedModel
 
 from vllm.attention.backends.abstract import AttentionMetadata
 from vllm.attention.layer import Attention
-from vllm.config import CacheConfig, VllmConfig
-from vllm.distributed import get_tensor_model_parallel_world_size
+from vllm.compilation.decorators import support_torch_compile
+from vllm.config import VllmConfig
+from vllm.distributed import divide, get_tensor_model_parallel_world_size
+from vllm.distributed.parallel_state import get_pp_group
 from vllm.forward_context import get_forward_context
+from vllm.model_executor.layers.activation import SiluAndMul
 from vllm.model_executor.layers.layernorm import RMSNorm
 from vllm.model_executor.layers.linear import (ColumnParallelLinear,
                                                MergedColumnParallelLinear,
                                                QKVParallelLinear,
                                                RowParallelLinear)
 from vllm.model_executor.layers.logits_processor import LogitsProcessor
+from vllm.model_executor.layers.mamba.mamba2_metadata import (
+    Mamba2Metadata, prepare_mamba2_metadata)
 from vllm.model_executor.layers.mamba.ops.causal_conv1d import (
     causal_conv1d_fn, causal_conv1d_update)
 from vllm.model_executor.layers.mamba.ops.mamba_ssm import (
-    selective_scan_fn, selective_state_update)
+    selective_state_update)
+from vllm.model_executor.layers.mamba.ops.ssd_combined import (
+    mamba_chunk_scan_combined)
 from vllm.model_executor.layers.quantization import QuantizationConfig
 from vllm.model_executor.layers.rotary_embedding import get_rope
+from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler
 from vllm.model_executor.layers.vocab_parallel_embedding import (
     DEFAULT_VOCAB_PADDING_SIZE, ParallelLMHead, VocabParallelEmbedding)
 from vllm.model_executor.model_loader.weight_utils import (
     composed_weight_loader, default_weight_loader, sharded_weight_loader)
 from vllm.model_executor.models.interfaces import (HasInnerState, IsHybrid,
-                                                   SupportsV0Only)
+                                                   SupportsPP, SupportsV0Only)
 from vllm.model_executor.models.mamba_cache import (MambaCacheManager,
                                                     MambaCacheParams)
-from vllm.model_executor.models.utils import maybe_prefix
+from vllm.model_executor.models.utils import (
+    is_pp_missing_parameter, make_empty_intermediate_tensors_factory,
+    make_layers, maybe_prefix)
 from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.model_executor.utils import set_weight_attrs
 from vllm.sequence import IntermediateTensors
@@ -77,17 +86,6 @@ class Plamo2PreTrainedModel(PreTrainedModel):  # type: ignore
                 module.weight.data[module.padding_idx].zero_()
 
 
-def get_initial_dt_bias(num_heads: int) -> torch.Tensor:
-    dt_min = 0.001
-    dt_max = 0.1
-    dt = torch.exp(
-        torch.rand(num_heads) * (math.log(dt_max) - math.log(dt_min)) +
-        math.log(dt_min))
-    dt = torch.clamp(dt, 1e-4)
-    inv_dt = dt + torch.log(-torch.expm1(-dt))
-    return inv_dt
-
-
 def is_mamba(config: Plamo2Config, i: int) -> bool:
     assert config.mamba_step > 1
 
@@ -97,52 +95,36 @@ def is_mamba(config: Plamo2Config, i: int) -> bool:
     return (i % config.mamba_step) != (config.mamba_step // 2)
 
 
-# TODO(Shinichi): Replace this with RMSNorm.
-def _rms_norm(hidden_states: torch.Tensor, weight: torch.Tensor,
-              eps: float) -> torch.Tensor:
-    input_shape = hidden_states.shape
-    hidden_states = hidden_states.reshape(input_shape[:-1] + weight.shape)
-    input_dtype = hidden_states.dtype
-    hidden_states = hidden_states.to(torch.float32)
-    variance = hidden_states.pow(2).mean(-1, keepdim=True)
-    hidden_states = hidden_states * torch.rsqrt(variance + eps)
-    hidden_states = hidden_states.to(input_dtype)
-    hidden_states = weight * hidden_states
-    return hidden_states.reshape(input_shape)
-
-
-def _swiglu(h: torch.Tensor) -> torch.Tensor:
-    h0, h1 = h.chunk(2, dim=-1)
-    return torch.nn.functional.silu(h0) * h1
-
-
-# Adapted from transformers.models.mamba.modeling_mamba.MambaMixer
+# Adapted from:
+# vllm.model_executor.layers.mamba.mamba_mixer2.MambaMixer2
+# transformers.models.mamba.modeling_mamba.MambaMixer
 class Plamo2MambaMixer(nn.Module):
-    # TODO(Shinichi): Rebase on Mamba2 implementation.
 
     def __init__(self,
-                 config: Plamo2Config,
-                 cache_config: CacheConfig,
-                 quant_config: QuantizationConfig,
-                 max_model_len: int,
+                 vllm_config: VllmConfig,
+                 *,
                  prefix: str = "",
                  **kwargs) -> None:
         super().__init__()
-        self.config = config
-        self.hidden_size = config.hidden_size
-        self.ssm_state_size = config.mamba_d_state
-        self.conv_kernel_size = config.mamba_d_conv
-        self.intermediate_size = (config.mamba_num_heads *
-                                  config.hidden_size_per_head)
-        self.hidden_size_per_head = config.hidden_size_per_head
-        self.num_heads = config.mamba_num_heads
+        self.config = vllm_config.model_config.hf_config
+        self.quant_config = vllm_config.quant_config
+        self.hidden_size = self.config.hidden_size
+        self.ssm_state_size = self.config.mamba_d_state
+        self.conv_kernel_size = self.config.mamba_d_conv
+        self.intermediate_size = (self.config.mamba_num_heads *
+                                  self.config.hidden_size_per_head)
+        self.tp_size = get_tensor_model_parallel_world_size()
+        self.intermediate_size_per_tp_worker = \
+            self.intermediate_size // self.tp_size
+        self.head_dim = self.config.hidden_size_per_head
+        self.num_heads = self.config.mamba_num_heads
         self.time_step_rank = max(64, self.hidden_size // 16)
-        self.use_conv_bias = False
-        self.use_bias = False
         self.conv1d = ColumnParallelLinear(
             input_size=self.conv_kernel_size,
             output_size=self.intermediate_size,
-            bias=self.use_conv_bias,
+            bias=False,
+            prefix=f"{prefix}.conv1d",
+            return_bias=False,
         )
         # unsqueeze to fit conv1d weights shape into the linear weights shape.
         # Can't do this in `weight_loader` since it already exists in
@@ -153,15 +135,19 @@ class Plamo2MambaMixer(nn.Module):
         self.in_proj = MergedColumnParallelLinear(
             self.hidden_size,
             [self.intermediate_size] * 2,
-            bias=self.use_bias,
+            bias=False,
+            quant_config=self.quant_config,
             prefix=f"{prefix}.in_proj",
+            return_bias=False,
         )
         # selective projection used to make dt, B and C input dependent
         self.bcdt_proj = RowParallelLinear(
             self.intermediate_size,
             self.time_step_rank + self.ssm_state_size * 2,
             bias=False,
+            quant_config=self.quant_config,
             prefix=f"{prefix}.bcdt_proj",
+            return_bias=False,
         )
         # time step projection (discretization) -
         # In the forward we need to apply dt_proj without the bias,
@@ -170,154 +156,224 @@ class Plamo2MambaMixer(nn.Module):
             self.time_step_rank,
             self.num_heads,
             bias=False,
+            quant_config=self.quant_config,
             prefix=f"{prefix}.dt_proj",
+            return_bias=False,
         )
-        self.dt_bias = torch.nn.Parameter(get_initial_dt_bias(self.num_heads))
 
-        tp_size = get_tensor_model_parallel_world_size()
         self.A = nn.Parameter(
             torch.empty(
-                self.intermediate_size // tp_size,
-                self.ssm_state_size,
+                divide(self.num_heads, self.tp_size),
                 dtype=torch.float32,
             ))
-        self.D = nn.Parameter(torch.ones(self.intermediate_size // tp_size))
+        self.D = nn.Parameter(torch.ones(divide(self.num_heads, self.tp_size)))
+        self.dt_bias = nn.Parameter(
+            torch.ones(divide(self.num_heads, self.tp_size)))
 
         set_weight_attrs(self.D, {"weight_loader": sharded_weight_loader(0)})
         a_weight_loader = composed_weight_loader(
             sharded_weight_loader(0), lambda x: -torch.exp(x.float()))
         set_weight_attrs(self.A, {"weight_loader": a_weight_loader})
+        set_weight_attrs(self.dt_bias,
+                         {"weight_loader": sharded_weight_loader(0)})
 
         self.out_proj = RowParallelLinear(
             self.intermediate_size,
             self.hidden_size,
-            bias=self.use_bias,
+            bias=False,
             input_is_parallel=True,
+            quant_config=self.quant_config,
             prefix=f"{prefix}.out_proj",
+            return_bias=False,
         )
         # The activation function is fixed to SiLU.
         self.activation = "silu"
 
-        self.dt_norm = RMSNorm(self.time_step_rank, eps=config.rms_norm_eps)
-        self.B_norm = RMSNorm(self.ssm_state_size, eps=config.rms_norm_eps)
-        self.C_norm = RMSNorm(self.ssm_state_size, eps=config.rms_norm_eps)
+        self.dt_norm = RMSNorm(self.time_step_rank,
+                               eps=self.config.rms_norm_eps)
+        self.B_norm = RMSNorm(self.ssm_state_size,
+                              eps=self.config.rms_norm_eps)
+        self.C_norm = RMSNorm(self.ssm_state_size,
+                              eps=self.config.rms_norm_eps)
 
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        mamba_cache_params: MambaCacheParams,
-        **kwargs,
-    ) -> torch.Tensor:
-
-        attn_metadata: AttentionMetadata = get_forward_context().attn_metadata
-
-        # 1. Gated MLP's linear projection
-        projected_states = self.in_proj(hidden_states)[0]
-        # Reshaping the projected states as in modeling_plamo.py.
-        length = len(hidden_states)
-        projected_states = projected_states.reshape(length, self.num_heads, -1)
-        gate, hidden_states = torch.split(
-            projected_states,
-            [self.hidden_size_per_head, self.hidden_size_per_head],
-            dim=-1)
-        hidden_states = hidden_states.reshape(length, -1).transpose(0, 1)
-        gate = gate.reshape(length, -1).transpose(0, 1)
-
-        # 2. Convolution sequence transformation
-        conv_weights = self.conv1d.weight.view(self.conv1d.weight.size(0),
-                                               self.conv1d.weight.size(2))
-
-        if attn_metadata.query_start_loc is not None \
-            and attn_metadata.context_lens_tensor is not None:
-            # |---------- N-1 iteration --------|
-            # |---------------- N iteration ---------------------|
-            # |- tokenA -|......................|-- newTokens ---|
-            # |---------- context_len ----------|
-            # |-------------------- seq_len ---------------------|
-            #                                   |-- query_len ---|
-            hidden_states = causal_conv1d_fn(
-                hidden_states,
-                conv_weights,
-                self.conv1d.bias,
-                activation=self.activation,
-                conv_states=mamba_cache_params.conv_state,
-                has_initial_state=attn_metadata.context_lens_tensor > 0,
-                cache_indices=mamba_cache_params.state_indices_tensor,
-                query_start_loc=attn_metadata.query_start_loc)
-        else:
-            hidden_states = causal_conv1d_update(
-                hidden_states.transpose(0, 1),
-                mamba_cache_params.conv_state,
-                conv_weights,
-                self.conv1d.bias,
-                self.activation,
-                conv_state_indices=mamba_cache_params.state_indices_tensor)
-            hidden_states = hidden_states.transpose(0, 1)
-
-        # 3. State Space Model sequence transformation
-        # 3.a. input varying initialization of time_step, B and C
-        ssm_parameters = self.bcdt_proj(hidden_states.transpose(-2, -1))[0]
-
-        # Splitting the ssm_parameters as in modeling_plamo.py.
+    def _project_ssm_parameters(self, hidden_states):
+        ssm_parameters = self.bcdt_proj(hidden_states)
         B, C, time_step = torch.split(
             ssm_parameters,
             [self.ssm_state_size, self.ssm_state_size, self.time_step_rank],
             dim=-1,
         )
+
+        # vllm._custom_ops.rms_norm requires contiguous input tensors.
         time_step = self.dt_norm(time_step.contiguous())
         B = self.B_norm(B.contiguous())
         C = self.C_norm(C.contiguous())
+        dt = self.dt_proj(time_step)
+        return B, C, dt
 
-        discrete_time_step = self.dt_proj(time_step)[0].transpose(-2, -1)
-        # 3.c perform the recurrence y ← SSM(A, B, C)(x)
-        time_proj_bias = (self.dt_bias.float() if hasattr(
-            self.dt_proj, "bias") else None)
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        mamba_cache_params: MambaCacheParams,
+        mamba2_metadata: Mamba2Metadata,
+        **kwargs,
+    ) -> torch.Tensor:
 
-        # Broadcasting as in modeling_plamo.py.
-        discrete_time_step = discrete_time_step.transpose(
-            0, 1)[..., None].expand(-1, -1, self.hidden_size_per_head)
-        discrete_time_step = discrete_time_step.reshape(
-            -1, self.intermediate_size).transpose(0, 1)
-        time_proj_bias = time_proj_bias[...,
-                                        None].expand(-1,
-                                                     self.hidden_size_per_head)
-        time_proj_bias = time_proj_bias.reshape(self.intermediate_size)
+        # mamba2_metadata contains metadata necessary for the mamba2 triton
+        # kernels to operate in continuous batching and in chunked prefill
+        # modes; they are computed at top-level model forward since they
+        # stay the same and reused for all mamba layers in the same iteration
+        attn_metadata: AttentionMetadata = get_forward_context().attn_metadata
 
-        if attn_metadata.query_start_loc is not None \
-            and attn_metadata.context_lens_tensor is not None:
-            scan_outputs = selective_scan_fn(
-                hidden_states,
-                mamba_cache_params.ssm_state,
-                discrete_time_step,
+        num_prefills = attn_metadata.num_prefills  # request count
+        num_decodes = attn_metadata.num_decode_tokens  # token count (=request)
+        num_prefill_tokens = attn_metadata.num_prefill_tokens  # token count
+        has_prefill = num_prefills > 0
+        has_decode = num_decodes > 0
+
+        # 1. Gated MLP's linear projection
+        projected_states = self.in_proj(hidden_states)
+        gate, hidden_states = projected_states.chunk(2, dim=-1)
+
+        # 2. Convolution sequence transformation
+        conv_weights = self.conv1d.weight.view(self.conv1d.weight.size(0),
+                                               self.conv1d.weight.size(2))
+
+        # Separate prefill and decode by splitting varlen input
+        # Split along token dimension
+        hidden_states_p, hidden_states_d = torch.split(
+            hidden_states,
+            [num_prefill_tokens, num_decodes],
+            dim=0,
+        )
+        gate_p, gate_d = torch.split(gate, [num_prefill_tokens, num_decodes],
+                                     dim=0)
+        # Split along batch dimension
+        state_indices_tensor_p, state_indices_tensor_d = torch.split(
+            mamba_cache_params.state_indices_tensor,
+            [num_prefills, num_decodes],
+            dim=0,
+        )
+        query_start_loc_p = (attn_metadata.query_start_loc[:num_prefills + 1]
+                             if has_prefill else None)
+
+        ssd_output_list = []
+
+        # Process prefill requests
+        if has_prefill:
+            # 2. Convolution sequence transformation
+            # - "cache_indices" updates the conv_state cache in positions
+            # pointed to by "mamba_cache_params.state_indices_tensor"
+            hidden_states_p = causal_conv1d_fn(
+                hidden_states_p.transpose(0, 1),
+                conv_weights,
+                self.conv1d.bias,
+                activation=self.activation,
+                conv_states=mamba_cache_params.conv_state,
+                has_initial_state=mamba2_metadata.has_initial_states,
+                cache_indices=state_indices_tensor_p,
+                query_start_loc=query_start_loc_p)
+            hidden_states_p = hidden_states_p.transpose(0, 1)
+            hidden_states_p = hidden_states_p[:num_prefill_tokens]
+            # In some instances, the following `bcdt_proj` op
+            # requires contiguous inputs
+            # (e.g. if the Marlin kernel is used).
+            hidden_states_p = hidden_states_p.contiguous()
+
+            B, C, dt = self._project_ssm_parameters(hidden_states_p)
+
+            # 3. State Space Model sequence transformation
+            initial_states = None
+            if (mamba2_metadata.has_initial_states is not None
+                    and mamba2_metadata.prep_initial_states):
+                # making a copy of the states
+                initial_states = torch.where(
+                    mamba2_metadata.has_initial_states[:, None, None, None],
+                    mamba_cache_params.ssm_state[state_indices_tensor_p], 0)
+            scan_output, varlen_state = mamba_chunk_scan_combined(
+                hidden_states_p.view(1, num_prefill_tokens,
+                                     self.num_heads // self.tp_size,
+                                     self.head_dim),
+                dt.unsqueeze(0),
                 self.A,
-                B.transpose(-2, -1),
-                C.transpose(-2, -1),
-                self.D.float(),
-                gate,
-                time_proj_bias,
-                delta_softplus=True,
-                cache_indices=mamba_cache_params.state_indices_tensor,
-                has_initial_state=attn_metadata.context_lens_tensor > 0,
-                query_start_loc=attn_metadata.query_start_loc)
-        else:
-            scan_outputs = selective_state_update(
+                B.view(1, num_prefill_tokens, 1, -1),
+                C.view(1, num_prefill_tokens, 1, -1),
+                chunk_size=mamba2_metadata.chunk_size,
+                D=self.D,
+                z=gate_p.view(1, num_prefill_tokens,
+                              self.num_heads // self.tp_size, self.head_dim),
+                dt_bias=self.dt_bias,
+                seq_idx=mamba2_metadata.seq_idx,
+                chunk_indices=mamba2_metadata.chunk_indices,
+                chunk_offsets=mamba2_metadata.chunk_offsets,
+                cu_seqlens=attn_metadata.query_start_loc[:num_prefills + 1],
+                initial_states=initial_states,
+                return_varlen_states=True,
+                return_final_states=False,
+                dt_softplus=True,
+                dt_limit=(0.0, float("inf")),
+            )
+
+            # update ssm states
+            # - varlen state is a (batch, nheads, headdim, dstate) tensor
+            mamba_cache_params.ssm_state[state_indices_tensor_p] = varlen_state
+
+            # - reshape
+            ssd_output_list.append(scan_output.view(num_prefill_tokens, -1))
+
+        # Process decode requests
+        if has_decode:
+            # 2. Convolution sequence transformation
+            hidden_states_d = causal_conv1d_update(
+                hidden_states_d,
+                mamba_cache_params.conv_state,
+                conv_weights,
+                self.conv1d.bias,
+                self.activation,
+                conv_state_indices=state_indices_tensor_d)
+
+            B, C, dt = self._project_ssm_parameters(hidden_states_d)
+
+            # 3. State Space Model sequence transformation
+            A = self.A[:, None, ...][:, :,
+                                     None].expand(-1, self.head_dim,
+                                                  self.config.mamba_d_state)
+            dt = dt[:, :, None].expand(-1, -1, self.head_dim)
+            dt_bias = self.dt_bias[:, None, ...].expand(-1, self.head_dim)
+            D = self.D[:, None, ...].expand(-1, self.head_dim)
+            B = B.unsqueeze(1)
+            C = C.unsqueeze(1)
+            hidden_states_d = hidden_states_d.view(
+                -1, self.num_heads // self.tp_size, self.head_dim)
+
+            # - the hidden is reshaped into (bs, num_heads, head_dim)
+            # - mamba_cache_params.ssm_state's slots will be selected
+            #   using state_indices_tensor_d
+
+            hidden_states_d = selective_state_update(
                 mamba_cache_params.ssm_state,
-                hidden_states.transpose(0, 1),
-                discrete_time_step.transpose(0, 1),
-                self.A,
+                hidden_states_d,
+                dt,
+                A,
                 B,
                 C,
-                self.D,
-                gate.transpose(0, 1),
-                time_proj_bias,
+                D,
+                z=gate_d.reshape(num_decodes, -1, self.head_dim),
+                dt_bias=dt_bias,
                 dt_softplus=True,
-                state_batch_indices=mamba_cache_params.state_indices_tensor)
-            scan_outputs = scan_outputs.transpose(0, 1)
+                state_batch_indices=state_indices_tensor_d,
+            )
+            assert self.num_heads % self.tp_size == 0
+            ssd_output_list.append(
+                hidden_states_d.view(-1, (self.num_heads // self.tp_size) *
+                                     self.head_dim))
+
+        # Merge prefill and decode outputs before passing to MLP
+        hidden_states = torch.vstack(ssd_output_list)
 
         # 4. Final linear projection
-        contextualized_states = self.out_proj(scan_outputs.transpose(-2,
-                                                                     -1))[0]
-        return contextualized_states
+        out = self.out_proj(hidden_states)
+        return out
 
 
 class DenseMLP(nn.Module):
@@ -332,33 +388,39 @@ class DenseMLP(nn.Module):
         self.hidden_size = config.hidden_size
         self.intermediate_size = config.intermediate_size
         self.gate_up_proj = MergedColumnParallelLinear(
-            self.hidden_size, [self.intermediate_size] * 2,
+            self.hidden_size,
+            [self.intermediate_size] * 2,
             bias=False,
             prefix=f"{prefix}.gate_up_proj",
-            quant_config=quant_config)
+            quant_config=quant_config,
+            return_bias=False,
+        )
+        self.act = SiluAndMul()
         self.down_proj = RowParallelLinear(self.intermediate_size,
                                            self.hidden_size,
                                            bias=False,
                                            prefix=f"{prefix}.down_proj",
-                                           quant_config=quant_config)
+                                           quant_config=quant_config,
+                                           return_bias=False)
 
     def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        h = self.gate_up_proj(hidden_states)[0]
-        h = _swiglu(h)
-        output, _ = self.down_proj(h)
-        return output  # type: ignore
+        h = self.gate_up_proj(hidden_states)
+        h = self.act(h)
+        return self.down_proj(h)
 
 
+@support_torch_compile
 class Plamo2AttentionMixer(nn.Module):
 
     def __init__(self,
-                 config: Plamo2Config,
-                 cache_config: CacheConfig,
-                 quant_config: QuantizationConfig,
-                 max_model_len: int | None = None,
+                 *,
+                 vllm_config: VllmConfig,
                  prefix: str = "",
                  **kwargs) -> None:
         super().__init__()
+        config = vllm_config.model_config.hf_config
+        cache_config = vllm_config.cache_config
+        quant_config = vllm_config.quant_config
         self.hidden_size = config.hidden_size
         tp_size = get_tensor_model_parallel_world_size()
         self.total_num_heads = config.num_attention_heads
@@ -396,19 +458,35 @@ class Plamo2AttentionMixer(nn.Module):
                                                        "rope_theta") else 10000
         self.rope_scaling = config.rope_scaling if hasattr(
             config, "rope_scaling") else None
+        max_position = config.max_position_embeddings
+        if hasattr(vllm_config.model_config, "max_model_len") and isinstance(
+                vllm_config.model_config.max_model_len, int):
+            max_position = min(max_position,
+                               vllm_config.model_config.max_model_len)
 
-        assert max_model_len is not None, "max_model_len must be provided"
         self.rotary_emb = get_rope(
             self.head_dim,
             rotary_dim=self.head_dim,
-            max_position=max_model_len,
+            max_position=max_position,
             base=self.rope_theta,
             rope_scaling=self.rope_scaling,
         )
-        self.q_weight = torch.nn.Parameter(
+        self.q_norm = RMSNorm(config.hidden_size_per_head,
+                              eps=config.rms_norm_eps)
+        self.q_norm.weight = torch.nn.Parameter(
             torch.ones((self.num_heads, config.hidden_size_per_head)))
-        self.k_weight = torch.nn.Parameter(
+        set_weight_attrs(self.q_norm.weight,
+                         {"weight_loader": sharded_weight_loader(0)})
+        self.k_norm = RMSNorm(config.hidden_size_per_head,
+                              eps=config.rms_norm_eps)
+        self.k_norm.weight = torch.nn.Parameter(
             torch.ones((self.num_kv_heads, config.hidden_size_per_head)))
+        # Tensor-parallelism shards the K norm weights to the tp ranks
+        # in a head-wise manner. This approach does not work if there is only
+        # a single KV head, as is the case for PLaMo 2-1B.
+        if self.total_num_kv_heads != 1:
+            set_weight_attrs(self.k_norm.weight,
+                             {"weight_loader": sharded_weight_loader(0)})
 
         self.attn = Attention(
             self.num_heads,
@@ -423,13 +501,18 @@ class Plamo2AttentionMixer(nn.Module):
         self,
         positions: torch.Tensor,
         hidden_states: torch.Tensor,
-        residual: Optional[torch.Tensor],
         **kwargs,
     ) -> torch.Tensor:
         qkv, _ = self.qkv_proj(hidden_states)
         q, k, v = qkv.split([self.q_size, self.kv_size, self.kv_size], dim=-1)
-        q = _rms_norm(q, self.q_weight, 1e-6)
-        k = _rms_norm(k, self.k_weight, 1e-6)
+
+        q_shape = q.shape
+        q = q.reshape(q_shape[:-1] + self.q_norm.weight.shape)
+        q = self.q_norm.forward_native(q).reshape(q_shape)
+        k_shape = k.shape
+        k = k.reshape(k_shape[:-1] + self.k_norm.weight.shape)
+        k = self.k_norm.forward_native(k).reshape(k_shape)
+
         q, k = self.rotary_emb(positions, q, k)
         attn_output = self.attn(q, k, v)
         output, _ = self.o_proj(attn_output)
@@ -441,27 +524,18 @@ class Plamo2DecoderLayer(nn.Module):
     def __init__(self,
                  vllm_config: VllmConfig,
                  layer_idx: int,
-                 max_model_len: int | None = None,
                  prefix: str = "",
                  **kwargs) -> None:
         super().__init__()
         config = vllm_config.model_config.hf_config
-        cache_config = vllm_config.cache_config
         quant_config = vllm_config.quant_config
-        max_model_len = vllm_config.scheduler_config.max_model_len
 
         self.is_mamba = is_mamba(config, layer_idx)
         if self.is_mamba:
-            self.mixer = Plamo2MambaMixer(config=config,
-                                          cache_config=cache_config,
-                                          quant_config=quant_config,
-                                          max_model_len=max_model_len,
+            self.mixer = Plamo2MambaMixer(vllm_config=vllm_config,
                                           prefix=f"{prefix}.mixer")
         else:
-            self.mixer = Plamo2AttentionMixer(config=config,
-                                              cache_config=cache_config,
-                                              quant_config=quant_config,
-                                              max_model_len=max_model_len,
+            self.mixer = Plamo2AttentionMixer(vllm_config=vllm_config,
                                               prefix=f"{prefix}.mixer")
 
         self.mlp = DenseMLP(config=config,
@@ -482,6 +556,7 @@ class Plamo2DecoderLayer(nn.Module):
         hidden_states: torch.Tensor,
         residual: Optional[torch.Tensor],
         mamba_cache_params: MambaCacheParams,
+        mamba2_metadata: Mamba2Metadata,
         **kwargs,
     ):
         if residual is None:
@@ -491,10 +566,12 @@ class Plamo2DecoderLayer(nn.Module):
             hidden_states, residual = self.pre_mixer_norm(
                 hidden_states, residual)
 
-        hidden_states = self.mixer(positions=positions,
-                                   hidden_states=hidden_states,
-                                   residual=residual,
-                                   mamba_cache_params=mamba_cache_params)
+        hidden_states = self.mixer(
+            positions=positions,
+            hidden_states=hidden_states,
+            mamba_cache_params=mamba_cache_params,
+            mamba2_metadata=mamba2_metadata,
+        )
         hidden_states = self.post_mixer_norm(hidden_states)
         # Fully Connected
         hidden_states, residual = self.pre_mlp_norm(hidden_states, residual)
@@ -507,14 +584,18 @@ class Plamo2Decoder(torch.nn.Module):
 
     def __init__(self, vllm_config: VllmConfig, prefix: str = "") -> None:
         super().__init__()
-        num_hidden_layers = vllm_config.model_config.hf_config.num_hidden_layers
+        config = vllm_config.model_config.hf_config
+        extra_kwargs = {"is_lora_enabled": bool(vllm_config.lora_config)}
 
-        self.layers = nn.ModuleList([
-            Plamo2DecoderLayer(vllm_config=vllm_config,
-                               layer_idx=i,
-                               prefix=f"{prefix}.layers.{i}")
-            for i in range(num_hidden_layers)
-        ])
+        def get_layer(prefix: str):
+            layer_idx = int(prefix.rsplit(".", 1)[1])
+            return Plamo2DecoderLayer(vllm_config=vllm_config,
+                                      layer_idx=layer_idx,
+                                      prefix=prefix,
+                                      **extra_kwargs)
+
+        self.start_layer, self.end_layer, self.layers = make_layers(
+            config.num_hidden_layers, get_layer, prefix=f"{prefix}.layers")
 
     def forward(
         self,
@@ -522,9 +603,10 @@ class Plamo2Decoder(torch.nn.Module):
         hidden_states: torch.Tensor,
         residual: Optional[torch.Tensor],
         mamba_cache_params: MambaCacheParams,
+        mamba2_metadata: Mamba2Metadata,
     ) -> torch.Tensor:
         mamba_cache_index = 0
-        for layer in self.layers:
+        for layer in self.layers[self.start_layer:self.end_layer]:
             layer_mamba_cache_params = None
             if layer.is_mamba:
                 layer_mamba_cache_params = mamba_cache_params.at_layer_idx(
@@ -535,7 +617,9 @@ class Plamo2Decoder(torch.nn.Module):
                 positions=positions,
                 hidden_states=hidden_states,
                 residual=residual,
-                mamba_cache_params=layer_mamba_cache_params)
+                mamba_cache_params=layer_mamba_cache_params,
+                mamba2_metadata=mamba2_metadata,
+            )
         return hidden_states, residual
 
 
@@ -557,10 +641,16 @@ class Plamo2Model(Plamo2PreTrainedModel):
             org_num_embeddings=config.vocab_size,
             prefix=f"{prefix}.embed_tokens",
         )
+        self.make_empty_intermediate_tensors = (
+            make_empty_intermediate_tensors_factory(
+                ["hidden_states", "residual"], config.hidden_size))
         self.layers = Plamo2Decoder(vllm_config, prefix=f"{prefix}.layers")
         self.norm = RMSNorm(config.hidden_size, eps=config.rms_norm_eps)
         self.post_init()
 
+    def get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor:
+        return self.embed_tokens(input_ids)
+
     def forward(
         self,
         input_ids: torch.Tensor,
@@ -569,21 +659,41 @@ class Plamo2Model(Plamo2PreTrainedModel):
         intermediate_tensors: Optional[IntermediateTensors] = None,
         inputs_embeds: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
-        # TODO(Shinichi): Implement pipeline parallelism.
-        hidden_states = self.embed_tokens(input_ids)
-        residual = None
+        if get_pp_group().is_first_rank:
+            if inputs_embeds is not None:
+                hidden_states = inputs_embeds
+            else:
+                hidden_states = self.get_input_embeddings(input_ids)
+            residual = None
+        else:
+            assert intermediate_tensors is not None
+            hidden_states = intermediate_tensors["hidden_states"]
+            residual = intermediate_tensors["residual"]
+
+        attn_metadata: AttentionMetadata = get_forward_context().attn_metadata
+        mamba2_metadata = prepare_mamba2_metadata(
+            chunk_size=self.config.mamba_chunk_size,
+            attn_metadata=attn_metadata,
+        )
 
         hidden_states, residual = self.layers(
             positions=positions,
             hidden_states=hidden_states,
             residual=residual,
-            mamba_cache_params=mamba_cache_params)
+            mamba_cache_params=mamba_cache_params,
+            mamba2_metadata=mamba2_metadata,
+        )
+        if not get_pp_group().is_last_rank:
+            return IntermediateTensors({
+                "hidden_states": hidden_states,
+                "residual": residual
+            })
         hidden_states, _ = self.norm(hidden_states, residual)
         return hidden_states
 
 
-class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
-                        SupportsV0Only):
+class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, SupportsPP,
+                        IsHybrid, SupportsV0Only):
     packed_modules_mapping = {
         "qkv_proj": [
             "q_proj",
@@ -629,10 +739,15 @@ class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
 
         self.logits_processor = LogitsProcessor(self.unpadded_vocab_size,
                                                 self.config.vocab_size)
-
+        self.sampler = get_sampler()
+        self.make_empty_intermediate_tensors = (
+            self.model.make_empty_intermediate_tensors)
         # Initialize weights and apply final processing
         self.post_init()
 
+    def get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor:
+        return self.model.get_input_embeddings(input_ids)
+
     def forward(self,
                 input_ids: torch.Tensor,
                 positions: torch.Tensor,
@@ -661,7 +776,7 @@ class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
         return self.mamba_cache.get_seqlen_agnostic_capture_inputs(batch_size)
 
     def _get_mamba_cache_shape(
-            self) -> tuple[tuple[int, int], tuple[int, int]]:
+            self) -> tuple[tuple[int, int], tuple[int, int, int]]:
         world_size = get_tensor_model_parallel_world_size()
         hidden_size = (self.config.mamba_num_heads *
                        self.config.hidden_size_per_head)
@@ -670,7 +785,8 @@ class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
             self.config.mamba_d_conv - 1,
         )
         temporal_state_shape = (
-            hidden_size // world_size,
+            divide(self.config.mamba_num_heads, world_size),
+            self.config.hidden_size_per_head,
             self.config.mamba_d_state,
         )
         return conv_state_shape, temporal_state_shape
@@ -684,6 +800,14 @@ class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
                                        sampling_metadata)
         return logits
 
+    def sample(
+        self,
+        logits: Optional[torch.Tensor],
+        sampling_metadata: SamplingMetadata,
+    ) -> Optional[SamplerOutput]:
+        next_tokens = self.sampler(logits, sampling_metadata)
+        return next_tokens
+
     def load_weights(self, weights: Iterable[tuple[str, torch.Tensor]]):
         params_dict = dict(self.named_parameters())
         for name, loaded_weight in weights:
@@ -703,23 +827,46 @@ class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
                 ".B_norm_weight": ".B_norm.weight",
                 ".C_norm_weight": ".C_norm.weight",
                 ".dt_norm_weight": ".dt_norm.weight",
+                ".q_weight": ".q_norm.weight",
+                ".k_weight": ".k_norm.weight",
             }
             # Apply replacements based on the defined mappings
             for old, new in replacements.items():
                 if old in name:
                     name = name.replace(old, new)
 
-            # Broadcast the loaded weight to match the model's parameter shape.
-            if ".A" in name:
-                loaded_weight = loaded_weight[:, None, None].expand(
-                    -1, self.config.hidden_size_per_head,
-                    self.config.mamba_d_state)
+            # Reshape the in_proj weights to match the shape expected
+            # by MergedColumnParallelLinear.
+            # This works both for unquantized weights and
+            # for quantized weights.
+            # In the quantized case, the weights are already transposed.
+            # Also, in addition to the quantized weights,
+            # the zero points and scales have to be reshaped as well.
+            # Packing should not be affected by this.
+            if ".mixer.in_proj.weight" in name \
+                or "mixer.in_proj.qweight" in name \
+                or "mixer.in_proj.scales" in name \
+                or "mixer.in_proj.qzeros" in name:
+                if "mixer.in_proj.weight" in name:
+                    loaded_weight = loaded_weight.transpose(0, 1)
+                # for weight:
+                # loaded_weight.shape[0] == self.config.hidden_size
+                # for qweight:
+                # loaded_weight.shape[0] == self.config.hidden_size // param.pack_factor  # noqa
+                # for scales and qzeros:
+                # loaded_weight.shape[0] == self.config.hidden_size // self.vllm_config.quant_config.group_size  # noqa
                 loaded_weight = loaded_weight.reshape(
-                    -1, self.config.mamba_d_state)
-            elif ".D" in name:
-                loaded_weight = loaded_weight[:, None].expand(
-                    -1, self.config.hidden_size_per_head)
-                loaded_weight = loaded_weight.reshape(-1)
+                    loaded_weight.shape[0], self.config.mamba_num_heads, -1)
+                gate_weight, hidden_states_weight = loaded_weight.chunk(2,
+                                                                        dim=-1)
+                gate_weight = gate_weight.reshape(loaded_weight.shape[0], -1)
+                hidden_states_weight = hidden_states_weight.reshape(
+                    loaded_weight.shape[0], -1)
+                loaded_weight = torch.cat([gate_weight, hidden_states_weight],
+                                          dim=-1)
+                if "mixer.in_proj.weight" in name:
+                    loaded_weight = loaded_weight.transpose(0, 1)
+
             # Offset parameter with vllm's RMSNorm haven't been supported yet.
             if ".pre_mixer_norm" in name:
                 loaded_weight += 1.0
@@ -732,6 +879,10 @@ class Plamo2ForCausalLM(Plamo2PreTrainedModel, HasInnerState, IsHybrid,
             elif "model.norm.weight" in name:
                 loaded_weight += 1.0
 
+            # Skip layers on other devices.
+            if is_pp_missing_parameter(name, self):
+                continue
+
             param = params_dict[name]
             weight_loader = getattr(param, "weight_loader",
                                     default_weight_loader)