[Model] Add support for OLMo architecture (#3046)

This PR add support for OLMo architecture.

Additional support: add support for clip-qkv.

Test: already tested on android(pixel 4) and cuda(setting tensor_parallel_shrads=2)

Author: Lanssi

python/mlc_llm/conversation_template/__init__.py

@@ -20,6 +20,7 @@
     llava,
     mistral,
     oasst,
+    olmo,
     orion,
     phi,
     qwen2,

python/mlc_llm/conversation_template/olmo.py

@@ -0,0 +1,28 @@
+"""OLMo default templates"""
+
+from mlc_llm.protocol.conversation_protocol import Conversation, MessagePlaceholders
+
+from .registry import ConvTemplateRegistry
+
+# Note that eos_token id is "50279" both in Allenai and AMD version.
+# So use the number instead of text.
+# Allenai version chat_template and eos_token:
+# https://huggingface.co/allenai/OLMo-7B-Instruct/blob/main/tokenizer_config.json
+# AMD version chat_template and eos_token:
+# https://huggingface.co/amd/AMD-OLMo-1B-SFT-DPO/blob/main/tokenizer_config.json
+ConvTemplateRegistry.register_conv_template(
+    Conversation(
+        name="olmo",
+        system_template=f"{MessagePlaceholders.SYSTEM.value}",
+        system_message="",
+        system_prefix_token_ids=[50279],
+        roles={
+            "user": "<|user|>",
+            "assistant": "<|assistant|>",
+        },
+        seps=["\n"],
+        role_content_sep="\n",
+        role_empty_sep="\n",
+        stop_token_ids=[50279],
+    )
+)

python/mlc_llm/interface/gen_config.py

@@ -306,4 +306,5 @@ def gen_config(  # pylint: disable=too-many-locals,too-many-arguments,too-many-b
     "aya-23",
     "deepseek_v2",
     "deepseek",
+    "olmo",
 }

python/mlc_llm/model/model.py

@@ -28,6 +28,7 @@
 from .minicpm import minicpm_loader, minicpm_model, minicpm_quantization
 from .mistral import mistral_loader, mistral_model, mistral_quantization
 from .mixtral import mixtral_loader, mixtral_model, mixtral_quantization
+from .olmo import olmo_loader, olmo_model, olmo_quantization
 from .orion import orion_loader, orion_model, orion_quantization
 from .phi import phi_loader, phi_model, phi_quantization
 from .phi3 import phi3_loader, phi3_model, phi3_quantization
@@ -532,4 +533,21 @@ class Model:
             "ft-quant": deepseek_quantization.ft_quant,
         },
     ),
+    "olmo": Model(
+        name="olmo",
+        model=olmo_model.OLMoForCausalLM,
+        config=olmo_model.OLMoConfig,
+        source={
+            "huggingface-torch": olmo_loader.huggingface,
+            "huggingface-safetensor": olmo_loader.huggingface,
+            "awq": olmo_loader.awq,
+        },
+        quantize={
+            "no-quant": olmo_quantization.no_quant,
+            "group-quant": olmo_quantization.group_quant,
+            "ft-quant": olmo_quantization.ft_quant,
+            "awq": olmo_quantization.awq_quant,
+            "per-tensor-quant": olmo_quantization.per_tensor_quant,
+        },
+    ),
 }

python/mlc_llm/model/olmo/__init__.py

@@ -0,0 +1,172 @@
+"""
+This file specifies how MLC's OLMo parameter maps from other formats, for example HuggingFace
+PyTorch, HuggingFace safetensors.
+"""
+
+import functools
+
+import numpy as np
+
+from mlc_llm.loader import ExternMapping
+from mlc_llm.quantization import Quantization
+
+from .olmo_model import OLMoConfig, OLMoForCausalLM
+from .olmo_quantization import awq_quant
+
+
+def huggingface(model_config: OLMoConfig, quantization: Quantization) -> ExternMapping:
+    """Returns a parameter mapping that maps from the names of MLC LLM parameters to
+    the names of HuggingFace PyTorch parameters.
+
+    Parameters
+    ----------
+    model_config : OLMoConfig
+        The configuration of the OLMo model.
+
+    quantization : Quantization
+        The quantization configuration.
+
+    Returns
+    -------
+    param_map : ExternMapping
+        The parameter mapping from MLC to HuggingFace PyTorch.
+    """
+    model = OLMoForCausalLM(model_config)
+    if quantization is not None:
+        model.to(quantization.model_dtype)
+    _, _named_params, _ = model.export_tvm(  # type: ignore[misc]
+        spec=model.get_default_spec(),
+        allow_extern=True,
+    )
+    named_parameters = dict(_named_params)
+
+    mapping = ExternMapping()
+
+    for i in range(model_config.num_hidden_layers):
+        # Add QKV in self attention
+        attn = f"model.layers.{i}.self_attn"
+        mlc_name = f"{attn}.qkv_proj.weight"
+        mlc_param = named_parameters[mlc_name]
+        mapping.add_mapping(
+            mlc_name,
+            [
+                f"{attn}.q_proj.weight",
+                f"{attn}.k_proj.weight",
+                f"{attn}.v_proj.weight",
+            ],
+            functools.partial(
+                lambda q, k, v, dtype: np.concatenate([q, k, v], axis=0).astype(dtype),
+                dtype=mlc_param.dtype,
+            ),
+        )
+        # Add gates in MLP
+        mlp = f"model.layers.{i}.mlp"
+        mlc_name = f"{mlp}.gate_up_proj.weight"
+        mlc_param = named_parameters[mlc_name]
+        mapping.add_mapping(
+            mlc_name,
+            [
+                f"{mlp}.gate_proj.weight",
+                f"{mlp}.up_proj.weight",
+            ],
+            functools.partial(
+                lambda gate, up, dtype: np.concatenate([gate, up], axis=0).astype(dtype),
+                dtype=mlc_param.dtype,
+            ),
+        )
+        # inv_freq is not used in the model
+        mapping.add_unused(f"{attn}.rotary_emb.inv_freq")
+
+    for mlc_name, mlc_param in named_parameters.items():
+        if mlc_name not in mapping.param_map:
+            mapping.add_mapping(
+                mlc_name,
+                [mlc_name],
+                functools.partial(
+                    lambda x, dtype: x.astype(dtype),
+                    dtype=mlc_param.dtype,
+                ),
+            )
+    return mapping
+
+
+def awq(model_config: OLMoConfig, quantization: Quantization) -> ExternMapping:
+    """Returns a parameter mapping that maps from the names of MLC LLM parameters to
+    the names of AWQ parameters.
+    Parameters
+    ----------
+    model_config : OLMoConfig
+        The configuration of the OLMo model.
+
+    quantization : Quantization
+        The quantization configuration.
+
+    Returns
+    -------
+    param_map : ExternMapping
+        The parameter mapping from MLC to AWQ.
+    """
+    model, _ = awq_quant(model_config, quantization)
+    _, _named_params, _ = model.export_tvm(  # type: ignore[misc]
+        spec=model.get_default_spec(),  # type: ignore[attr-defined]
+        allow_extern=True,
+    )
+    named_parameters = dict(_named_params)
+
+    mapping = ExternMapping()
+
+    for i in range(model_config.num_hidden_layers):
+        # Add QKV in self attention
+        attn = f"model.layers.{i}.self_attn"
+        for quantize_suffix in ["qweight", "qzeros", "scales"]:
+            mlc_name = f"{attn}.qkv_proj.{quantize_suffix}"
+            assert mlc_name in named_parameters
+            mlc_param = named_parameters[mlc_name]
+            mapping.add_mapping(
+                mlc_name,
+                [
+                    f"{attn}.q_proj.{quantize_suffix}",
+                    f"{attn}.k_proj.{quantize_suffix}",
+                    f"{attn}.v_proj.{quantize_suffix}",
+                ],
+                functools.partial(
+                    lambda q, k, v, dtype: np.concatenate(
+                        [q, k, v],
+                        axis=1,  # AWQ GEMM would transpose the weight
+                    ).astype(dtype),
+                    dtype=mlc_param.dtype,
+                ),
+            )
+
+        # Concat gate and up in MLP
+        mlp = f"model.layers.{i}.mlp"
+        for quantize_suffix in ["qweight", "qzeros", "scales"]:
+            mlc_name = f"{mlp}.gate_up_proj.{quantize_suffix}"
+            assert mlc_name in named_parameters
+            mlc_param = named_parameters[mlc_name]
+            mapping.add_mapping(
+                mlc_name,
+                [
+                    f"{mlp}.gate_proj.{quantize_suffix}",
+                    f"{mlp}.up_proj.{quantize_suffix}",
+                ],
+                functools.partial(
+                    lambda gate, up, dtype: np.concatenate(
+                        [gate, up],
+                        axis=1,  # AWQ GEMM would transpose the weight
+                    ).astype(dtype),
+                    dtype=mlc_param.dtype,
+                ),
+            )
+
+        # inv_freq is not used in the model
+        mapping.add_unused(f"{attn}.rotary_emb.inv_freq")
+
+    for mlc_name, mlc_param in named_parameters.items():
+        if mlc_name not in mapping.param_map:
+            mapping.add_mapping(
+                mlc_name,
+                [mlc_name],
+                functools.partial(lambda x, dtype: x.astype(dtype), dtype=mlc_param.dtype),
+            )
+    return mapping