Initial commit

llama_toolchain/inference/__init__.py

@@ -0,0 +1,5 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.

llama_toolchain/inference/api/__init__.py

@@ -0,0 +1,8 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from .datatypes import *  # noqa: F401 F403
+from .endpoints import *  # noqa: F401 F403

llama_toolchain/inference/api/config.py

@@ -0,0 +1,94 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from enum import Enum
+from typing import Literal, Optional, Union
+
+from hydra.core.config_store import ConfigStore
+
+from hydra_zen import builds
+from llama_models.llama3_1.api.datatypes import CheckpointQuantizationFormat
+
+from pydantic import BaseModel, Field
+from strong_typing.schema import json_schema_type
+from typing_extensions import Annotated
+
+from .datatypes import QuantizationConfig
+
+
+@json_schema_type
+class ImplType(Enum):
+    inline = "inline"
+    remote = "remote"
+
+
+@json_schema_type
+class CheckpointType(Enum):
+    pytorch = "pytorch"
+    huggingface = "huggingface"
+
+
+@json_schema_type
+class PytorchCheckpoint(BaseModel):
+    checkpoint_type: Literal[CheckpointType.pytorch.value] = (
+        CheckpointType.pytorch.value
+    )
+    checkpoint_dir: str
+    tokenizer_path: str
+    model_parallel_size: int
+    quantization_format: CheckpointQuantizationFormat = (
+        CheckpointQuantizationFormat.bf16
+    )
+
+
+@json_schema_type
+class HuggingFaceCheckpoint(BaseModel):
+    checkpoint_type: Literal[CheckpointType.huggingface.value] = (
+        CheckpointType.huggingface.value
+    )
+    repo_id: str  # or model_name ?
+    model_parallel_size: int
+    quantization_format: CheckpointQuantizationFormat = (
+        CheckpointQuantizationFormat.bf16
+    )
+
+
+@json_schema_type
+class ModelCheckpointConfig(BaseModel):
+    checkpoint: Annotated[
+        Union[PytorchCheckpoint, HuggingFaceCheckpoint],
+        Field(discriminator="checkpoint_type"),
+    ]
+
+
+@json_schema_type
+class InlineImplConfig(BaseModel):
+    impl_type: Literal[ImplType.inline.value] = ImplType.inline.value
+    checkpoint_config: ModelCheckpointConfig
+    quantization: Optional[QuantizationConfig] = None
+    torch_seed: Optional[int] = None
+    max_seq_len: int
+    max_batch_size: int = 1
+
+
+@json_schema_type
+class RemoteImplConfig(BaseModel):
+    impl_type: Literal[ImplType.remote.value] = ImplType.remote.value
+    url: str = Field(..., description="The URL of the remote module")
+
+
+@json_schema_type
+class InferenceConfig(BaseModel):
+    impl_config: Annotated[
+        Union[InlineImplConfig, RemoteImplConfig],
+        Field(discriminator="impl_type"),
+    ]
+
+
+InferenceHydraConfig = builds(InferenceConfig)
+
+cs = ConfigStore.instance()
+cs.store(name="inference_config", node=InferenceHydraConfig)

llama_toolchain/inference/api/datatypes.py

@@ -0,0 +1,72 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from enum import Enum
+from typing import List, Literal, Optional, Union
+
+from pydantic import BaseModel, Field
+
+from strong_typing.schema import json_schema_type
+from typing_extensions import Annotated
+
+from llama_models.llama3_1.api.datatypes import *  # noqa: F403
+
+
+class LogProbConfig(BaseModel):
+    top_k: Optional[int] = 0
+
+
+@json_schema_type
+class QuantizationType(Enum):
+    bf16 = "bf16"
+    fp8 = "fp8"
+
+
+@json_schema_type
+class Fp8QuantizationConfig(BaseModel):
+    type: Literal[QuantizationType.fp8.value] = QuantizationType.fp8.value
+
+
+@json_schema_type
+class Bf16QuantizationConfig(BaseModel):
+    type: Literal[QuantizationType.bf16.value] = QuantizationType.bf16.value
+
+
+QuantizationConfig = Annotated[
+    Union[Bf16QuantizationConfig, Fp8QuantizationConfig],
+    Field(discriminator="type"),
+]
+
+
+@json_schema_type
+class ChatCompletionResponseEventType(Enum):
+    start = "start"
+    complete = "complete"
+    progress = "progress"
+
+
+@json_schema_type
+class ToolCallParseStatus(Enum):
+    started = "started"
+    in_progress = "in_progress"
+    failure = "failure"
+    success = "success"
+
+
+@json_schema_type
+class ToolCallDelta(BaseModel):
+    content: Union[str, ToolCall]
+    parse_status: ToolCallParseStatus
+
+
+@json_schema_type
+class ChatCompletionResponseEvent(BaseModel):
+    """Chat completion response event."""
+
+    event_type: ChatCompletionResponseEventType
+    delta: Union[str, ToolCallDelta]
+    logprobs: Optional[List[TokenLogProbs]] = None
+    stop_reason: Optional[StopReason] = None

llama_toolchain/inference/api/endpoints.py

@@ -0,0 +1,123 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from .datatypes import *  # noqa: F403
+from typing import Optional, Protocol
+
+# this dependency is annoying and we need a forked up version anyway
+from pyopenapi import webmethod
+
+
+@json_schema_type
+class CompletionRequest(BaseModel):
+    model: PretrainedModel
+    content: InterleavedTextAttachment
+    sampling_params: Optional[SamplingParams] = SamplingParams()
+
+    stream: Optional[bool] = False
+    logprobs: Optional[LogProbConfig] = None
+    quantization_config: Optional[QuantizationConfig] = None
+
+
+@json_schema_type
+class CompletionResponse(BaseModel):
+    completion_message: CompletionMessage
+    logprobs: Optional[List[TokenLogProbs]] = None
+
+
+@json_schema_type
+class CompletionResponseStreamChunk(BaseModel):
+    """streamed completion response."""
+
+    delta: str
+    stop_reason: Optional[StopReason] = None
+    logprobs: Optional[List[TokenLogProbs]] = None
+
+
+@json_schema_type
+class BatchCompletionRequest(BaseModel):
+    model: PretrainedModel
+    content_batch: List[InterleavedTextAttachment]
+    sampling_params: Optional[SamplingParams] = SamplingParams()
+    logprobs: Optional[LogProbConfig] = None
+    quantization_config: Optional[QuantizationConfig] = None
+
+
+@json_schema_type
+class BatchCompletionResponse(BaseModel):
+    completion_message_batch: List[CompletionMessage]
+
+
+@json_schema_type
+class ChatCompletionRequest(BaseModel):
+    model: InstructModel
+    messages: List[Message]
+    sampling_params: Optional[SamplingParams] = SamplingParams()
+
+    # zero-shot tool definitions as input to the model
+    available_tools: Optional[List[ToolDefinition]] = Field(default_factory=list)
+
+    stream: Optional[bool] = False
+    logprobs: Optional[LogProbConfig] = None
+    quantization_config: Optional[QuantizationConfig] = None
+
+
+@json_schema_type
+class ChatCompletionResponseStreamChunk(BaseModel):
+    """SSE-stream of these events."""
+
+    event: ChatCompletionResponseEvent
+
+
+@json_schema_type
+class ChatCompletionResponse(BaseModel):
+    completion_message: CompletionMessage
+    logprobs: Optional[List[TokenLogProbs]] = None
+
+
+@json_schema_type
+class BatchChatCompletionRequest(BaseModel):
+    model: InstructModel
+    messages_batch: List[List[Message]]
+    sampling_params: Optional[SamplingParams] = SamplingParams()
+
+    # zero-shot tool definitions as input to the model
+    available_tools: Optional[List[ToolDefinition]] = Field(default_factory=list)
+
+    logprobs: Optional[LogProbConfig] = None
+    quantization_config: Optional[QuantizationConfig] = None
+
+
+@json_schema_type
+class BatchChatCompletionResponse(BaseModel):
+    completion_message_batch: List[CompletionMessage]
+
+
+class Inference(Protocol):
+
+    @webmethod(route="/inference/completion")
+    async def completion(
+        self,
+        request: CompletionRequest,
+    ) -> Union[CompletionResponse, CompletionResponseStreamChunk]: ...
+
+    @webmethod(route="/inference/chat_completion")
+    async def chat_completion(
+        self,
+        request: ChatCompletionRequest,
+    ) -> Union[ChatCompletionResponse, ChatCompletionResponseStreamChunk]: ...
+
+    @webmethod(route="/inference/batch_completion")
+    async def batch_completion(
+        self,
+        request: BatchCompletionRequest,
+    ) -> BatchCompletionResponse: ...
+
+    @webmethod(route="/inference/batch_chat_completion")
+    async def batch_chat_completion(
+        self,
+        request: BatchChatCompletionRequest,
+    ) -> BatchChatCompletionResponse: ...

llama_toolchain/inference/api_instance.py

@@ -0,0 +1,18 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from .api.config import ImplType, InferenceConfig
+
+
+async def get_inference_api_instance(config: InferenceConfig):
+    if config.impl_config.impl_type == ImplType.inline.value:
+        from .inference import InferenceImpl
+
+        return InferenceImpl(config.impl_config)
+
+    from .client import InferenceClient
+
+    return InferenceClient(config.impl_config.url)

llama_toolchain/inference/client.py

@@ -0,0 +1,85 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+import asyncio
+import json
+from typing import AsyncGenerator
+
+import fire
+import httpx
+from termcolor import cprint
+
+from .api import (
+    ChatCompletionRequest,
+    ChatCompletionResponseStreamChunk,
+    CompletionRequest,
+    Inference,
+    InstructModel,
+    UserMessage,
+)
+from .event_logger import EventLogger
+
+
+class InferenceClient(Inference):
+    def __init__(self, base_url: str):
+        print(f"Initializing client for {base_url}")
+        self.base_url = base_url
+
+    async def initialize(self) -> None:
+        pass
+
+    async def shutdown(self) -> None:
+        pass
+
+    async def completion(self, request: CompletionRequest) -> AsyncGenerator:
+        raise NotImplementedError()
+
+    async def chat_completion(self, request: ChatCompletionRequest) -> AsyncGenerator:
+        async with httpx.AsyncClient() as client:
+            async with client.stream(
+                "POST",
+                f"{self.base_url}/inference/chat_completion",
+                data=request.json(),
+                headers={"Content-Type": "application/json"},
+                timeout=20,
+            ) as response:
+                async for line in response.aiter_lines():
+                    if line.startswith("data:"):
+                        data = line[len("data: ") :]
+                        try:
+                            yield ChatCompletionResponseStreamChunk(**json.loads(data))
+                        except Exception as e:
+                            print(data)
+                            print(f"Error with parsing or validation: {e}")
+
+
+async def run_main(host: str, port: int):
+    client = InferenceClient(f"http://{host}:{port}")
+
+    message = UserMessage(content="hello world, help me out here")
+    cprint(f"User>{message.content}", "green")
+    req = ChatCompletionRequest(
+        model=InstructModel.llama3_70b_chat,
+        messages=[message],
+        stream=True,
+    )
+    iterator = client.chat_completion(
+        ChatCompletionRequest(
+            model=InstructModel.llama3_8b_chat,
+            messages=[message],
+            stream=True,
+        )
+    )
+    async for log in EventLogger().log(iterator):
+        log.print()
+
+
+def main(host: str, port: int):
+    asyncio.run(run_main(host, port))
+
+
+if __name__ == "__main__":
+    fire.Fire(main)

llama_toolchain/inference/event_logger.py

@@ -0,0 +1,36 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from termcolor import cprint
+
+from llama_toolchain.inference.api import ChatCompletionResponseEventType
+
+
+class LogEvent:
+    def __init__(
+        self,
+        content: str = "",
+        end: str = "\n",
+        color="white",
+    ):
+        self.content = content
+        self.color = color
+        self.end = "\n" if end is None else end
+
+    def print(self, flush=True):
+        cprint(f"{self.content}", color=self.color, end=self.end, flush=flush)
+
+
+class EventLogger:
+    async def log(self, event_generator, stream=True):
+        async for chunk in event_generator:
+            event = chunk.event
+            if event.event_type == ChatCompletionResponseEventType.start:
+                yield LogEvent("Assistant> ", color="cyan", end="")
+            elif event.event_type == ChatCompletionResponseEventType.progress:
+                yield LogEvent(event.delta, color="yellow", end="")
+            elif event.event_type == ChatCompletionResponseEventType.complete:
+                yield LogEvent("")

llama_toolchain/inference/generation.py

@@ -0,0 +1,319 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# This software may be used and distributed in accordance with the terms of the Llama 3 Community License Agreement.
+
+import json
+import os
+import sys
+import time
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Generator, List, Optional
+
+import torch
+import torch.nn.functional as F
+from fairscale.nn.model_parallel.initialize import (
+    get_model_parallel_rank,
+    initialize_model_parallel,
+    model_parallel_is_initialized,
+)
+from llama_models.llama3_1.api.args import ModelArgs
+from llama_models.llama3_1.api.chat_format import ChatFormat, ModelInput
+from llama_models.llama3_1.api.datatypes import Message
+from llama_models.llama3_1.api.model import Transformer
+from llama_models.llama3_1.api.tokenizer import Tokenizer
+from termcolor import cprint
+
+from .api.config import CheckpointType, InlineImplConfig
+from .api.datatypes import QuantizationType
+
+
+@dataclass
+class TokenResult:
+    token: int
+    text: str
+    logprobs: Optional[List[float]] = None
+
+
+class Llama:
+    @staticmethod
+    def build(config: InlineImplConfig):
+        """
+        Build a Llama instance by initializing and loading a model checkpoint.
+
+        Note:
+            This method initializes the distributed process group, sets the device to CUDA,
+            and loads the pre-trained model and tokenizer.
+        """
+        checkpoint = config.checkpoint_config.checkpoint
+        if checkpoint.checkpoint_type != CheckpointType.pytorch.value:
+            raise NotImplementedError("HuggingFace checkpoints not supported yet")
+
+        if (
+            config.quantization
+            and config.quantization.type == QuantizationType.fp8.value
+        ):
+            from .quantization.loader import is_fbgemm_available
+
+            if not is_fbgemm_available():
+                raise ImportError("fbgemm-gpu is required for FP8 quantization")
+
+        if not torch.distributed.is_initialized():
+            torch.distributed.init_process_group("nccl")
+
+        model_parallel_size = checkpoint.model_parallel_size
+        if not model_parallel_is_initialized():
+            initialize_model_parallel(model_parallel_size)
+
+        local_rank = int(os.environ.get("LOCAL_RANK", 0))
+        torch.cuda.set_device(local_rank)
+
+        # seed must be the same in all processes
+        if config.torch_seed is not None:
+            torch.manual_seed(config.torch_seed)
+
+        if local_rank > 0:
+            sys.stdout = open(os.devnull, "w")
+
+        start_time = time.time()
+        ckpt_dir = checkpoint.checkpoint_dir
+        checkpoints = sorted(Path(ckpt_dir).glob("*.pth"))
+        assert len(checkpoints) > 0, f"no checkpoint files found in {ckpt_dir}"
+        assert model_parallel_size == len(
+            checkpoints
+        ), f"Loading a checkpoint for MP={len(checkpoints)} but world size is {model_parallel_size}"
+        ckpt_path = checkpoints[get_model_parallel_rank()]
+        state_dict = torch.load(ckpt_path, map_location="cpu", weights_only=True)
+        with open(Path(ckpt_dir) / "params.json", "r") as f:
+            params = json.loads(f.read())
+
+        # TODO(ashwin): this block is so we can load internal checkpoints without additional
+        # fuss. the final code should _not_ have this blurb
+        if "model" in params:
+            params = params["model"]
+
+        model_args: ModelArgs = ModelArgs(
+            max_seq_len=config.max_seq_len,
+            max_batch_size=config.max_batch_size,
+            **params,
+        )
+        tokenizer = Tokenizer(model_path=checkpoint.tokenizer_path)
+
+        assert (
+            model_args.vocab_size == tokenizer.n_words
+        ), f"model_args vocab = {model_args.vocab_size} but tokenizer vocab = {tokenizer.n_words}"
+
+        fp8 = (
+            config.quantization
+            and config.quantization.type == QuantizationType.fp8.value
+        )
+
+        if fp8:
+            # load on CPU in bf16 so that fp8 conversion does not find an
+            # unexpected (fp32, e.g.) datatype
+            torch.set_default_tensor_type(torch.BFloat16Tensor)
+
+        model = Transformer(model_args)
+
+        if fp8:
+            # load on CPU first since if we are doing fp8, we probably don't
+            # have enough memory on GPU for bf16
+            model.load_state_dict(state_dict, strict=False)
+
+        if torch.cuda.is_bf16_supported():
+            torch.set_default_tensor_type(torch.cuda.BFloat16Tensor)
+        else:
+            torch.set_default_tensor_type(torch.cuda.HalfTensor)
+
+        if not fp8:
+            model.load_state_dict(state_dict, strict=False)
+
+        if config.quantization:
+            from .quantization.loader import convert_to_quantized_model
+
+            model = convert_to_quantized_model(model, config)
+        else:
+            model = model.to("cuda")
+
+        print(f"Loaded in {time.time() - start_time:.2f} seconds")
+
+        return Llama(model, tokenizer, model_args)
+
+    def __init__(self, model: Transformer, tokenizer: Tokenizer, args: ModelArgs):
+        self.args = args
+        self.model = model
+        self.tokenizer = tokenizer
+        self.formatter = ChatFormat(tokenizer)
+
+    @torch.inference_mode()
+    def generate(
+        self,
+        model_input: ModelInput,
+        max_gen_len: int,
+        temperature: float = 0.6,
+        top_p: float = 0.9,
+        logprobs: bool = False,
+        echo: bool = False,
+        include_stop_token: bool = False,
+    ) -> Generator:
+        params = self.model.params
+
+        # cprint("Input to model -> " + self.tokenizer.decode(model_input.tokens), "red")
+        prompt_tokens = [model_input.tokens]
+
+        bsz = 1
+        assert bsz <= params.max_batch_size, (bsz, params.max_batch_size)
+
+        min_prompt_len = min(len(t) for t in prompt_tokens)
+        max_prompt_len = max(len(t) for t in prompt_tokens)
+
+        if max_prompt_len >= params.max_seq_len:
+            cprint(
+                f"Out of token budget {max_prompt_len} vs {params.max_seq_len}", "red"
+            )
+            return
+
+        total_len = min(max_gen_len + max_prompt_len, params.max_seq_len)
+        pad_id = self.tokenizer.pad_id
+        tokens = torch.full((bsz, total_len), pad_id, dtype=torch.long, device="cuda")
+        for k, t in enumerate(prompt_tokens):
+            tokens[k, : len(t)] = torch.tensor(t, dtype=torch.long, device="cuda")
+        if logprobs:
+            token_logprobs = torch.zeros_like(tokens, dtype=torch.float)
+
+        prev_pos = 0
+        eos_reached = torch.tensor([False] * bsz, device="cuda")
+        input_text_mask = tokens != pad_id
+        if min_prompt_len == total_len:
+            # TODO(ashwin): unify this branch with the one below and figure out multimodal crap
+            logits = self.model.forward(tokens, prev_pos)
+            token_logprobs = -F.cross_entropy(
+                input=logits.transpose(1, 2),
+                target=tokens,
+                reduction="none",
+                ignore_index=pad_id,
+            )
+
+        stop_tokens = torch.tensor(self.tokenizer.stop_tokens)
+
+        for cur_pos in range(min_prompt_len, total_len):
+            logits = self.model.forward(tokens[:, prev_pos:cur_pos], prev_pos)
+
+            if temperature > 0:
+                probs = torch.softmax(logits[:, -1] / temperature, dim=-1)
+                next_token = sample_top_p(probs, top_p)
+            else:
+                next_token = torch.argmax(logits[:, -1], dim=-1)
+
+            next_token = next_token.reshape(-1)
+            # only replace token if prompt has already been generated
+            next_token = torch.where(
+                input_text_mask[:, cur_pos], tokens[:, cur_pos], next_token
+            )
+            tokens[:, cur_pos] = next_token
+
+            target = tokens[:, prev_pos + 1 : cur_pos + 1]
+            if logprobs:
+                token_logprobs[:, prev_pos + 1 : cur_pos + 1] = -F.cross_entropy(
+                    input=logits.transpose(1, 2),
+                    target=tokens[:, prev_pos + 1 : cur_pos + 1],
+                    reduction="none",
+                    ignore_index=pad_id,
+                )
+            eos_reached |= (~input_text_mask[:, cur_pos]) & (
+                torch.isin(next_token, stop_tokens)
+            )
+            yield TokenResult(
+                token=next_token[0].item(),
+                text=self.tokenizer.decode(next_token.tolist()),
+                logprobs=(
+                    token_logprobs[:, prev_pos + 1 : cur_pos + 1][0].tolist()
+                    if logprobs
+                    else None
+                ),
+            )
+
+            prev_pos = cur_pos
+            if all(eos_reached):
+                break
+
+    def text_completion(
+        self,
+        prompt: str,
+        temperature: float = 0.6,
+        top_p: float = 0.9,
+        max_gen_len: Optional[int] = None,
+        logprobs: bool = False,
+        echo: bool = False,
+    ) -> Generator:
+        if (
+            max_gen_len is None
+            or max_gen_len == 0
+            or max_gen_len >= self.model.params.max_seq_len
+        ):
+            max_gen_len = self.model.params.max_seq_len - 1
+
+        prompt_tokens = self.tokenizer.encode(x, bos=True, eos=False)
+
+        yield from self.generate(
+            model_input=ModelInput(tokens=prompt_tokens),
+            max_gen_len=max_gen_len,
+            temperature=temperature,
+            top_p=top_p,
+            logprobs=logprobs,
+            echo=echo,
+        )
+
+    def chat_completion(
+        self,
+        messages: List[Message],
+        temperature: float = 0.6,
+        top_p: float = 0.9,
+        max_gen_len: Optional[int] = None,
+        logprobs: bool = False,
+    ) -> Generator:
+        if (
+            max_gen_len is None
+            or max_gen_len == 0
+            or max_gen_len >= self.model.params.max_seq_len
+        ):
+            max_gen_len = self.model.params.max_seq_len - 1
+
+        yield from self.generate(
+            model_input=self.formatter.encode_dialog_prompt(messages),
+            max_gen_len=max_gen_len,
+            temperature=temperature,
+            top_p=top_p,
+            logprobs=logprobs,
+            include_stop_token=True,
+        )
+
+
+def sample_top_p(probs, p):
+    """
+    Perform top-p (nucleus) sampling on a probability distribution.
+
+    Args:
+        probs (torch.Tensor): Probability distribution tensor.
+        p (float): Probability threshold for top-p sampling.
+
+    Returns:
+        torch.Tensor: Sampled token indices.
+
+    Note:
+        Top-p sampling selects the smallest set of tokens whose cumulative probability mass
+        exceeds the threshold p. The distribution is renormalized based on the selected tokens.
+    """
+    probs_sort, probs_idx = torch.sort(probs, dim=-1, descending=True)
+    probs_sum = torch.cumsum(probs_sort, dim=-1)
+    mask = probs_sum - probs_sort > p
+    probs_sort[mask] = 0.0
+    probs_sort.div_(probs_sort.sum(dim=-1, keepdim=True))
+    next_token = torch.multinomial(probs_sort, num_samples=1)
+    next_token = torch.gather(probs_idx, -1, next_token)
+    return next_token

llama_toolchain/inference/inference.py

@@ -0,0 +1,159 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from typing import AsyncGenerator
+
+from llama_models.llama3_1.api.datatypes import StopReason
+
+from .api.config import InlineImplConfig
+from .api.datatypes import (
+    ChatCompletionResponseEvent,
+    ChatCompletionResponseEventType,
+    ToolCallDelta,
+    ToolCallParseStatus,
+)
+from .api.endpoints import (
+    ChatCompletionRequest,
+    ChatCompletionResponseStreamChunk,
+    CompletionRequest,
+    Inference,
+)
+from .model_parallel import LlamaModelParallelGenerator
+
+
+class InferenceImpl(Inference):
+
+    def __init__(self, config: InlineImplConfig) -> None:
+        self.config = config
+
+    async def initialize(self) -> None:
+        self.generator = LlamaModelParallelGenerator(self.config)
+        self.generator.start()
+
+    async def shutdown(self) -> None:
+        self.generator.stop()
+
+    async def completion(self, request: CompletionRequest) -> AsyncGenerator:
+        raise NotImplementedError()
+
+    async def chat_completion(self, request: ChatCompletionRequest) -> AsyncGenerator:
+        yield ChatCompletionResponseStreamChunk(
+            event=ChatCompletionResponseEvent(
+                event_type=ChatCompletionResponseEventType.start,
+                delta="",
+            )
+        )
+
+        tokens = []
+        logprobs = []
+
+        stop_reason = None
+
+        buffer = ""
+        ipython = False
+
+        for token_result in self.generator.chat_completion(
+            messages=request.messages,
+            temperature=request.sampling_params.temperature,
+            top_p=request.sampling_params.top_p,
+            max_gen_len=request.sampling_params.max_tokens,
+            logprobs=request.logprobs,
+        ):
+            buffer += token_result.text
+            tokens.append(token_result.token)
+
+            if not ipython and buffer.startswith("<|python_tag|>"):
+                ipython = True
+                yield ChatCompletionResponseStreamChunk(
+                    event=ChatCompletionResponseEvent(
+                        event_type=ChatCompletionResponseEventType.progress,
+                        delta=ToolCallDelta(
+                            content="",
+                            parse_status=ToolCallParseStatus.started,
+                        ),
+                    )
+                )
+                buffer = buffer[len("<|python_tag|>") :]
+                continue
+
+            if not request.stream:
+                if request.logprobs:
+                    logprobs.append(token_result.logprob)
+
+                continue
+
+            if token_result.text == "<|eot_id|>":
+                stop_reason = StopReason.end_of_turn
+                text = ""
+            elif token_result.text == "<|eom_id|>":
+                stop_reason = StopReason.end_of_message
+                text = ""
+            else:
+                text = token_result.text
+
+            if ipython:
+                delta = ToolCallDelta(
+                    content=text,
+                    parse_status=ToolCallParseStatus.in_progress,
+                )
+            else:
+                delta = text
+            yield ChatCompletionResponseStreamChunk(
+                event=ChatCompletionResponseEvent(
+                    event_type=ChatCompletionResponseEventType.progress,
+                    delta=delta,
+                    stop_reason=stop_reason,
+                )
+            )
+
+        if stop_reason is None:
+            stop_reason = StopReason.out_of_tokens
+
+        # TODO(ashwin): parse tool calls separately here and report errors?
+        # if someone breaks the iteration before coming here we are toast
+        message = self.generator.formatter.decode_assistant_message(tokens, stop_reason)
+        if request.stream:
+            parsed_tool_calls = len(message.tool_calls) > 0
+            if ipython and not parsed_tool_calls:
+                yield ChatCompletionResponseStreamChunk(
+                    event=ChatCompletionResponseEvent(
+                        event_type=ChatCompletionResponseEventType.progress,
+                        delta=ToolCallDelta(
+                            content="",
+                            parse_status=ToolCallParseStatus.failure,
+                        ),
+                        stop_reason=stop_reason,
+                    )
+                )
+
+            for tool_call in message.tool_calls:
+                yield ChatCompletionResponseStreamChunk(
+                    event=ChatCompletionResponseEvent(
+                        event_type=ChatCompletionResponseEventType.progress,
+                        delta=ToolCallDelta(
+                            content=tool_call,
+                            parse_status=ToolCallParseStatus.success,
+                        ),
+                        stop_reason=stop_reason,
+                    )
+                )
+
+            yield ChatCompletionResponseStreamChunk(
+                event=ChatCompletionResponseEvent(
+                    event_type=ChatCompletionResponseEventType.complete,
+                    delta="",
+                    stop_reason=stop_reason,
+                )
+            )
+
+            # TODO(ashwin): what else do we need to send out here when everything finishes?
+        else:
+            yield ChatCompletionResponse(
+                content=message.content,
+                tool_calls=message.tool_calls,
+                stop_reason=stop_reason,
+                logprobs=logprobs if request.logprobs else None,
+            )

llama_toolchain/inference/model_parallel.py

@@ -0,0 +1,104 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from copy import deepcopy
+from dataclasses import dataclass
+from functools import partial
+from typing import Generator, List, Optional
+
+from llama_models.llama3_1.api.chat_format import ChatFormat
+from llama_models.llama3_1.api.datatypes import Message
+from llama_models.llama3_1.api.tokenizer import Tokenizer
+
+from .api.config import InlineImplConfig
+from .generation import Llama
+from .parallel_utils import ModelParallelProcessGroup
+
+
+@dataclass
+class InferenceArgs:
+    messages: List[Message]
+    temperature: float
+    top_p: float
+    max_gen_len: int
+    logprobs: bool
+
+
+class ModelRunner:
+    def __init__(self, llama):
+        self.llama = llama
+
+    # the `task` object is the same that is sent to `ModelParallelProcessGroup.run_inference()`
+    def __call__(self, task: InferenceArgs):
+        return self.llama.chat_completion(
+            task.messages,
+            task.temperature,
+            task.top_p,
+            task.max_gen_len,
+            task.logprobs,
+        )
+
+
+def init_model_cb(config: InlineImplConfig):
+    llama = Llama.build(config)
+    return ModelRunner(llama)
+
+
+class LlamaModelParallelGenerator:
+    """
+    This abstraction exists so
+     - we can run model parallel code without needing to run the CLIs via torchrun
+     - this also enables use model parallel code within a notebook context.
+
+    A Context Manager is used to ensure that the model parallel process is started and stopped
+    correctly. This does make the ergonomics a little awkward, because it isn't immediately
+    clear at the callsite why we need to use a context manager.
+    """
+
+    def __init__(self, config: InlineImplConfig):
+        self.config = config
+
+        # this is a hack because Agent's loop uses this to tokenize and check if input is too long
+        # while the tool-use loop is going
+        checkpoint = self.config.checkpoint_config.checkpoint
+        self.formatter = ChatFormat(Tokenizer(checkpoint.tokenizer_path))
+
+    def start(self):
+        self.__enter__()
+
+    def stop(self):
+        self.__exit__(None, None, None)
+
+    def __enter__(self):
+        checkpoint = self.config.checkpoint_config.checkpoint
+        self.group = ModelParallelProcessGroup(
+            checkpoint.model_parallel_size,
+            init_model_cb=partial(init_model_cb, self.config),
+        )
+        self.group.start()
+        return self
+
+    def __exit__(self, exc_type, exc_value, exc_traceback):
+        self.group.stop()
+
+    def chat_completion(
+        self,
+        messages: List[Message],
+        temperature: float = 0.6,
+        top_p: float = 0.9,
+        max_gen_len: Optional[int] = None,
+        logprobs: bool = False,
+    ) -> Generator:
+        req_obj = InferenceArgs(
+            messages=deepcopy(messages),
+            temperature=temperature,
+            top_p=top_p,
+            max_gen_len=max_gen_len,
+            logprobs=logprobs,
+        )
+
+        gen = self.group.run_inference(req_obj)
+        yield from gen

llama_toolchain/inference/parallel_utils.py

@@ -0,0 +1,265 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+import multiprocessing
+import os
+import pickle
+import tempfile
+import time
+import uuid
+
+from typing import Callable, Generator
+
+import torch
+
+import zmq
+
+from fairscale.nn.model_parallel.initialize import (
+    get_model_parallel_group,
+    get_model_parallel_rank,
+    get_model_parallel_src_rank,
+)
+
+from torch.distributed.launcher.api import elastic_launch, LaunchConfig
+
+
+_END_SENTINEL = "__end_sentinel__"
+_CANCEL_SENTINEL = "__cancel_sentinel__"
+
+
+def mp_rank_0() -> bool:
+    return get_model_parallel_rank() == 0
+
+
+def retrieve_requests(reply_socket_url: str):
+    if mp_rank_0():
+        context = zmq.Context()
+        reply_socket = context.socket(zmq.ROUTER)
+        reply_socket.connect(reply_socket_url)
+
+        while True:
+            client_id, obj = maybe_get_work(reply_socket)
+            if obj is None:
+                time.sleep(0.01)
+                continue
+
+            reply_socket.send_multipart([client_id, pickle.dumps("YES READY")])
+            break
+
+    def send_obj(obj):
+        reply_socket.send_multipart([client_id, pickle.dumps(obj)])
+
+    while True:
+        tasks = [None]
+        if mp_rank_0():
+            client_id, task = maybe_get_work(reply_socket)
+            # there is still an unknown unclean GeneratorExit happening resulting in a
+            # cancel sentinel getting queued _after_ we have finished sending everything :/
+            # kind of a hack this is :/
+            if task != _CANCEL_SENTINEL:
+                tasks = [task]
+
+        torch.distributed.broadcast_object_list(
+            tasks,
+            src=get_model_parallel_src_rank(),
+            group=get_model_parallel_group(),
+        )
+
+        task = tasks[0]
+        if task is None:
+            time.sleep(0.1)
+        else:
+            try:
+                out = yield task
+                if out is None:
+                    break
+
+                for obj in out:
+                    updates = [None]
+                    if mp_rank_0():
+                        _, update = maybe_get_work(reply_socket)
+                        if update == _CANCEL_SENTINEL:
+                            updates = [update]
+                        else:
+                            # only send the update if it's not cancelled otherwise the object sits in the socket
+                            # and gets pulled in the next request lol
+                            send_obj(obj)
+
+                    torch.distributed.broadcast_object_list(
+                        updates,
+                        src=get_model_parallel_src_rank(),
+                        group=get_model_parallel_group(),
+                    )
+                    if updates[0] == _CANCEL_SENTINEL:
+                        print("quitting generation loop because request was cancelled")
+                        break
+
+                if mp_rank_0():
+                    send_obj(_END_SENTINEL)
+            except Exception as e:
+                print(f"[debug] got exception {e}")
+                import traceback
+
+                traceback.print_exc()
+                if mp_rank_0():
+                    send_obj(e)
+
+    if mp_rank_0():
+        send_obj("DONE")
+
+
+def maybe_get_work(sock: zmq.Socket):
+    message = None
+    client_id = None
+    try:
+        client_id, obj = sock.recv_multipart(zmq.NOBLOCK)
+        message = pickle.loads(obj)
+    except zmq.ZMQError as e:
+        if e.errno != zmq.EAGAIN:
+            raise e
+
+    return client_id, message
+
+
+def worker_process_entrypoint(
+    reply_socket_url: str,
+    init_model_cb: Callable,
+) -> None:
+    model = init_model_cb()
+    torch.distributed.barrier()
+    time.sleep(1)
+
+    # run the requests co-routine which retrieves requests from the socket
+    # and sends responses (we provide) back to the caller
+    req_gen = retrieve_requests(reply_socket_url)
+    result = None
+    while True:
+        try:
+            task = req_gen.send(result)
+            if isinstance(task, str) and task == _END_SENTINEL:
+                break
+
+            result = model(task)
+        except StopIteration:
+            break
+
+    print("[debug] worker process done")
+
+
+def launch_dist_group(
+    reply_socket_url: str,
+    model_parallel_size: int,
+    init_model_cb: Callable,
+    **kwargs,
+) -> None:
+    id = uuid.uuid4().hex
+    dist_url = f"file:///tmp/llama3_{id}_{time.time()}"
+
+    with tempfile.TemporaryDirectory() as tmpdir:
+        # TODO: track workers and if they terminate, tell parent process about it so cleanup can happen
+        launch_config = LaunchConfig(
+            max_nodes=1,
+            min_nodes=1,
+            nproc_per_node=model_parallel_size,
+            start_method="fork",
+            rdzv_backend="c10d",
+            rdzv_endpoint=os.path.join(tmpdir, "rdzv"),
+            rdzv_configs={"store_type": "file", "timeout": 90},
+            max_restarts=0,
+            monitor_interval=1,
+            run_id=str(uuid.uuid4()),
+        )
+        elastic_launch(launch_config, entrypoint=worker_process_entrypoint)(
+            reply_socket_url,
+            init_model_cb,
+        )
+
+
+def start_model_parallel_process(
+    model_parallel_size: int,
+    init_model_cb: Callable,
+    **kwargs,
+):
+    context = zmq.Context()
+    request_socket = context.socket(zmq.DEALER)
+
+    # Binding the request socket to a random port
+    request_socket.bind("tcp://127.0.0.1:0")
+
+    main_process_url = request_socket.getsockopt_string(zmq.LAST_ENDPOINT)
+
+    ctx = multiprocessing.get_context("fork")
+    process = ctx.Process(
+        target=launch_dist_group,
+        args=(
+            main_process_url,
+            model_parallel_size,
+            init_model_cb,
+        ),
+        kwargs=kwargs,
+    )
+    process.start()
+
+    # wait until the model is loaded; rank 0 will send a message to indicate it's ready
+
+    request_socket.send_pyobj("READY?")
+    response = request_socket.recv_pyobj()
+    print(f"Finished model load {response}")
+
+    return request_socket, process
+
+
+class ModelParallelProcessGroup:
+    def __init__(
+        self,
+        model_parallel_size: int,
+        init_model_cb: Callable,
+        **kwargs,
+    ):
+        self.model_parallel_size = model_parallel_size
+        self.init_model_cb = init_model_cb
+        self.started = False
+        self.running = False
+
+    def start(self):
+        assert not self.started, "process group already started"
+        self.request_socket, self.process = start_model_parallel_process(
+            self.model_parallel_size,
+            self.init_model_cb,
+        )
+        self.started = True
+
+    def stop(self):
+        assert self.started, "process group not started"
+        if self.process.is_alive():
+            self.request_socket.send_pyobj(_END_SENTINEL, zmq.NOBLOCK)
+            self.process.join()
+        self.started = False
+
+    def run_inference(self, request) -> Generator:
+        assert not self.running, "inference already running"
+
+        self.running = True
+        self.request_socket.send_pyobj(request)
+        try:
+            while True:
+                obj = self.request_socket.recv_pyobj()
+                if obj == _END_SENTINEL:
+                    break
+
+                if isinstance(obj, Exception):
+                    print(f"[debug] got exception {obj}")
+                    raise obj
+
+                yield obj
+        except GeneratorExit as e:
+            self.request_socket.send_pyobj(_CANCEL_SENTINEL)
+            while True:
+                obj = self.request_socket.recv_pyobj()
+                if obj == _END_SENTINEL:
+                    break
+        finally:
+            self.running = False

llama_toolchain/inference/quantization/fp8_impls.py

@@ -0,0 +1,184 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# This software may be used and distributed in accordance with the terms of the Llama 3 Community License Agreement.
+
+import collections
+from typing import Optional, Type
+
+try:
+    import fbgemm_gpu.experimental.gen_ai  # noqa: F401
+
+    print("Using efficient FP8 operators in FBGEMM.")
+except ImportError:
+    print("No efficient FP8 operators. Please install FBGEMM in fp8_requirements.txt.")
+    raise
+
+import torch
+from torch import nn, Tensor
+
+
+class Fp8ScaledWeights:
+    # TODO: Ugly trick so torch allows us to replace parameters
+    # with our custom Fp8Weights instance. Do this properly.
+    @property
+    def __class__(self) -> Type[nn.parameter.Parameter]:
+        return nn.Parameter
+
+    @property
+    def grad_fn(self) -> None:
+        return None
+
+
+# pyre-fixme[4]: Attribute annotation cannot be `Any`.
+# pyre-fixme[2]: Parameter annotation cannot be `Any`.
+class Fp8RowwiseWeights(
+    Fp8ScaledWeights,
+    collections.namedtuple(
+        "Fp8RowwiseWeights",
+        ["weight", "scale", "shape", "activation_scale_ub"],
+    ),
+):
+    pass
+
+
+def ffn_swiglu(
+    x: Tensor,
+    w1: Fp8RowwiseWeights,
+    w3: Fp8RowwiseWeights,
+    w2: Fp8RowwiseWeights,
+    num_tokens: Optional[Tensor] = None,
+    is_memory_bounded: bool = False,
+) -> Tensor:
+    if (
+        isinstance(w1, Fp8ScaledWeights)
+        and isinstance(w3, Fp8ScaledWeights)
+        and isinstance(w2, Fp8ScaledWeights)
+    ):
+        return ffn_swiglu_fp8_dynamic(
+            x, w1, w3, w2, w1.activation_scale_ub, num_tokens, is_memory_bounded
+        )
+
+    (B, T, D) = x.shape  # noqa: N806
+    (HD_L, D_) = w1.shape  # noqa: N806
+    assert D_ == D
+
+    assert isinstance(w1, Tensor)
+    assert isinstance(w3, Tensor)
+    x1 = x.view(B * T, D) @ w1.T
+    x2 = x.view(B * T, D) @ w3.T
+    z = torch.nn.functional.silu(x1) * x2
+    del x1, x2
+    assert isinstance(w2, Tensor)
+    return (z @ w2.T).view(B, T, D)
+
+
+@torch.inference_mode()
+def quantize_fp8(
+    w: Tensor,
+    fp8_activation_scale_ub: float,
+    output_device: Optional[torch.device] = None,
+) -> Fp8RowwiseWeights:
+    """Quantize [n, k] weight tensor.
+
+    Args:
+        w (Tensor): [n, k] input high precision tensor to quantize.
+        fp8_activation_scale_ub (float): Upper bound for activation max.
+    """
+    activation_scale_ub = torch.tensor(
+        [fp8_activation_scale_ub],
+        dtype=torch.float,
+        device="cuda",
+    )
+    wq, w_scale = torch.ops.fbgemm.quantize_fp8_per_row(w)
+    del w
+    return Fp8RowwiseWeights(
+        weight=wq,
+        scale=w_scale,
+        shape=wq.shape,
+        activation_scale_ub=activation_scale_ub,
+    )
+
+
+@torch.inference_mode()
+def load_fp8(
+    w: Tensor,
+    w_scale: Tensor,
+    fp8_activation_scale_ub: float,
+) -> Fp8RowwiseWeights:
+    """Load FP8 [n, k] weight tensor.
+
+    Args:
+        w (Tensor): [n, k] input FP8.
+        fp8_activation_scale_ub (float): Upper bound for activation max.
+    """
+    activation_scale_ub = torch.tensor(
+        [fp8_activation_scale_ub],
+        dtype=torch.float,
+        device="cuda",
+    )
+    return Fp8RowwiseWeights(
+        weight=w.to(torch.float8_e4m3fn).to(device="cuda"),
+        scale=w_scale.to(device="cuda"),
+        shape=w.shape,
+        activation_scale_ub=activation_scale_ub,
+    )
+
+
+def fc_fp8_dynamic(
+    x: Tensor,
+    w: Fp8RowwiseWeights,
+    activation_scale_ub: Optional[Tensor] = None,
+    num_tokens: Optional[Tensor] = None,
+    is_memory_bounded: bool = False,
+) -> Tensor:
+    """
+    Single w8a8 fc layer with dynamic row-wise scaling.
+    """
+    if isinstance(w, Fp8RowwiseWeights):
+        xq, x_scale = torch.ops.fbgemm.quantize_fp8_per_row(
+            x, num_tokens, activation_scale_ub
+        )
+        y = torch.ops.fbgemm.f8f8bf16_rowwise(
+            xq, w.weight, x_scale, w.scale, use_fast_accum=True
+        )
+    del xq
+    return y
+
+
+def ffn_swiglu_fp8_dynamic(
+    x: Tensor,
+    w1: Fp8RowwiseWeights,
+    w3: Fp8RowwiseWeights,
+    w2: Fp8RowwiseWeights,
+    activation_scale_ub: Optional[Tensor] = None,
+    num_tokens: Optional[Tensor] = None,
+    is_memory_bounded: bool = False,
+) -> Tensor:
+    (B, T, D) = x.shape  # noqa: N806
+    HD_L = w1.shape[0]  # noqa: N806
+    assert HD_L == w3.shape[0]
+    x1 = fc_fp8_dynamic(
+        x.view(B * T, D),
+        w1,
+        activation_scale_ub,
+        num_tokens,
+        is_memory_bounded,
+    )
+    x2 = fc_fp8_dynamic(
+        x.view(B * T, D),
+        w3,
+        activation_scale_ub,
+        num_tokens,
+        is_memory_bounded,
+    )
+    z = torch.nn.functional.silu(x1) * x2
+    del x1, x2
+
+    z_ = fc_fp8_dynamic(z, w2, activation_scale_ub, num_tokens, is_memory_bounded)
+
+    return z_.view(B, T, D)

llama_toolchain/inference/quantization/loader.py

@@ -0,0 +1,105 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# This software may be used and distributed in accordance with the terms of the Llama 3 Community License Agreement.
+
+import os
+from typing import Optional
+
+import torch
+
+from fairscale.nn.model_parallel.mappings import reduce_from_model_parallel_region
+from llama_models.llama3_1.api.model import Transformer, TransformerBlock
+
+from llama_toolchain.inference.api.config import (
+    CheckpointQuantizationFormat,
+    InlineImplConfig,
+)
+from llama_toolchain.inference.api.datatypes import QuantizationType
+
+from termcolor import cprint
+from torch import Tensor
+
+
+def is_fbgemm_available() -> bool:
+    try:
+        import fbgemm_gpu.experimental.gen_ai  # noqa: F401
+
+        return True
+    except ImportError:
+        return False
+
+
+def swiglu_wrapper(
+    self,
+    x: Tensor,
+):
+    from .fp8_impls import ffn_swiglu
+
+    out = ffn_swiglu(x, self.w1.weight, self.w3.weight, self.w2.weight)
+    return reduce_from_model_parallel_region(out)
+
+
+def convert_to_quantized_model(
+    model: Transformer,
+    config: InlineImplConfig,
+    fp8_activation_scale_ub: Optional[float] = 1200.0,
+) -> Transformer:
+    if config.quantization.type == QuantizationType.bf16.value:
+        return model
+
+    elif config.quantization.type != QuantizationType.fp8.value:
+        raise ValueError("Only FP8 quantization is supported")
+
+    from .fp8_impls import Fp8ScaledWeights, load_fp8, quantize_fp8
+
+    checkpoint = config.checkpoint_config.checkpoint
+    # Move weights to GPU with quantization
+    if checkpoint.quantization_format == CheckpointQuantizationFormat.fp8_mixed.value:
+        cprint("Loading fp8 scales...", "yellow")
+        fp8_scales_path = os.path.join(
+            checkpoint.checkpoint_dir, f"fp8_scales_{get_model_parallel_rank()}.pt"
+        )
+        assert os.path.isfile(
+            fp8_scales_path
+        ), f"fp8_scales_path not found for rank {get_model_parallel_rank()}"
+        fp8_scales = torch.load(fp8_scales_path, weights_only=True)
+
+        for block in model.layers:
+            if isinstance(block, TransformerBlock):
+                if block.layer_id == 0 or block.layer_id == (model.n_layers - 1):
+                    continue
+
+                block.feed_forward.forward = swiglu_wrapper.__get__(block.feed_forward)
+                for key in ("w1", "w3", "w2"):
+                    param = getattr(block.feed_forward, key)
+                    param.weight = load_fp8(
+                        param.weight,
+                        fp8_scales[
+                            f"{block.layer_id}_feed_forward.{key}_{get_model_parallel_rank()}"
+                        ],
+                        fp8_activation_scale_ub,
+                    )
+    else:
+        cprint("Quantizing fp8 weights from bf16...", "yellow")
+        for block in model.layers:
+            if isinstance(block, TransformerBlock):
+                if block.layer_id == 0 or block.layer_id == (model.n_layers - 1):
+                    continue
+                block.feed_forward.forward = swiglu_wrapper.__get__(block.feed_forward)
+                for key in ("w1", "w3", "w2"):
+                    param = getattr(block.feed_forward, key)
+                    param.weight = quantize_fp8(
+                        param.weight,
+                        fp8_activation_scale_ub,
+                        output_device=torch.device("cuda"),
+                    )
+
+    for _, parameter in model.named_parameters():
+        if not isinstance(parameter, Fp8ScaledWeights):
+            parameter.data = parameter.to(device="cuda")
+    return model

llama_toolchain/inference/quantization/scripts/build_conda.sh

@@ -0,0 +1,30 @@
+#!/bin/bash
+
+if [[ $# -ne 1 ]]; then
+    echo "Error: Please provide the name of CONDA environment you wish to create"
+    exit 1
+fi
+
+ENV_NAME=$1
+
+set -eu
+eval "$(conda shell.bash hook)"
+
+echo "Will build env (or overwrite) named '$ENV_NAME'"
+
+set -x
+
+run_build() {
+    # Set up the conda environment
+    yes | conda remove --name $ENV_NAME --all
+    yes | conda create -n $ENV_NAME python=3.10
+    conda activate $ENV_NAME
+
+    # PT nightly
+    pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu121
+
+    # install dependencies for `llama-agentic-system`
+    pip install -r fp8_requirements.txt
+}
+
+run_build

llama_toolchain/inference/quantization/scripts/quantize_checkpoint.py

@@ -0,0 +1,161 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# This software may be used and distributed in accordance with the terms of the Llama 3 Community License Agreement.
+
+import json
+import os
+import shutil
+import sys
+from pathlib import Path
+from typing import Optional
+
+import fire
+
+import torch
+from fairscale.nn.model_parallel.initialize import (
+    get_model_parallel_rank,
+    initialize_model_parallel,
+    model_parallel_is_initialized,
+)
+from fp8.fp8_impls import FfnQuantizeMode, quantize_fp8
+
+from llama.model import ModelArgs, Transformer, TransformerBlock
+from llama.tokenizer import Tokenizer
+from torch.nn.parameter import Parameter
+
+
+def main(
+    ckpt_dir: str,
+    tokenizer_path: str,
+    quantized_ckpt_dir: str,
+    max_seq_len: Optional[int] = 512,
+    max_batch_size: Optional[int] = 4,
+    model_parallel_size: Optional[int] = None,
+    ffn_quantize_mode: Optional[FfnQuantizeMode] = FfnQuantizeMode.FP8_ROWWISE,
+    fp8_activation_scale_ub: Optional[float] = 1200.0,
+    seed: int = 1,
+):
+    """ """
+    if not os.path.exists(quantized_ckpt_dir):
+        os.makedirs(quantized_ckpt_dir)
+        shutil.copy(
+            os.path.join(ckpt_dir, "params.json"),
+            os.path.join(quantized_ckpt_dir, "params.json"),
+        )
+        shutil.copy(
+            os.path.join(ckpt_dir, "tokenizer.model"),
+            os.path.join(quantized_ckpt_dir, "tokenizer.model"),
+        )
+
+    if not torch.distributed.is_initialized():
+        torch.distributed.init_process_group("nccl")
+        if not model_parallel_is_initialized():
+            if model_parallel_size is None:
+                model_parallel_size = int(os.environ.get("WORLD_SIZE", 1))
+            initialize_model_parallel(model_parallel_size)
+
+        local_rank = int(os.environ.get("LOCAL_RANK", 0))
+        torch.cuda.set_device(local_rank)
+
+        # seed must be the same in all processes
+        torch.manual_seed(seed)
+
+        if local_rank > 0:
+            sys.stdout = open(os.devnull, "w")
+
+        checkpoints = sorted(Path(ckpt_dir).glob("*.pth"))
+        assert len(checkpoints) > 0, f"no checkpoint files found in {ckpt_dir}"
+        assert model_parallel_size == len(
+            checkpoints
+        ), f"Loading a checkpoint for MP={len(checkpoints)} but world size is {model_parallel_size}"
+        ckpt_path = checkpoints[get_model_parallel_rank()]
+        checkpoint = torch.load(ckpt_path, map_location="cpu", weights_only=True)
+        with open(Path(ckpt_dir) / "params.json", "r") as f:
+            params = json.loads(f.read())
+
+        model_args: ModelArgs = ModelArgs(
+            max_seq_len=max_seq_len,
+            max_batch_size=max_batch_size,
+            **params,
+        )
+        tokenizer = Tokenizer(model_path=tokenizer_path)
+        assert (
+            model_args.vocab_size == tokenizer.n_words
+        ), f"model_args vocab = {model_args.vocab_size} but tokenizer vocab = {tokenizer.n_words}"
+
+        # load on CPU in bf16 so that fp8 conversion does not find an unexpected (fp32, e.g.) datatype
+        torch.set_default_tensor_type(torch.BFloat16Tensor)
+
+        model = Transformer(model_args)
+        model.load_state_dict(checkpoint, strict=False)
+
+        if torch.cuda.is_bf16_supported():
+            torch.set_default_tensor_type(torch.cuda.BFloat16Tensor)
+        else:
+            torch.set_default_tensor_type(torch.cuda.HalfTensor)
+
+        print(ckpt_path)
+        assert (
+            quantized_ckpt_dir is not None
+        ), "QUantized checkpoint directory should not be None"
+        fp8_scales = {}
+        for block in model.layers:
+            if isinstance(block, TransformerBlock):
+                if block.layer_id == 0 or block.layer_id == (model.n_layers - 1):
+                    continue
+
+                fp8_weight = quantize_fp8(
+                    block.feed_forward.w1.weight,
+                    fp8_activation_scale_ub,
+                    ffn_quantize_mode,
+                    output_device=torch.device("cpu"),
+                )
+                with torch.inference_mode():
+                    block.feed_forward.w1.weight = Parameter(fp8_weight.weight)
+                fp8_scales[
+                    f"{block.layer_id}_feed_forward.w1_{get_model_parallel_rank()}"
+                ] = fp8_weight.scale
+
+                fp8_weight = quantize_fp8(
+                    block.feed_forward.w3.weight,
+                    fp8_activation_scale_ub,
+                    ffn_quantize_mode,
+                    output_device=torch.device("cpu"),
+                )
+                with torch.inference_mode():
+                    block.feed_forward.w3.weight = Parameter(fp8_weight.weight)
+                fp8_scales[
+                    f"{block.layer_id}_feed_forward.w3_{get_model_parallel_rank()}"
+                ] = fp8_weight.scale
+
+                fp8_weight = quantize_fp8(
+                    block.feed_forward.w2.weight,
+                    fp8_activation_scale_ub,
+                    ffn_quantize_mode,
+                    output_device=torch.device("cpu"),
+                )
+                with torch.inference_mode():
+                    block.feed_forward.w2.weight = Parameter(fp8_weight.weight)
+                fp8_scales[
+                    f"{block.layer_id}_feed_forward.w2_{get_model_parallel_rank()}"
+                ] = fp8_weight.scale
+
+        fp8_scales_path = os.path.join(
+            quantized_ckpt_dir, f"fp8_scales_{get_model_parallel_rank()}.pt"
+        )
+        torch.save(fp8_scales, fp8_scales_path)
+
+        ckpt_path = os.path.join(
+            quantized_ckpt_dir,
+            "consolidated.{:02d}.pth".format(get_model_parallel_rank()),
+        )
+        torch.save(model.state_dict(), ckpt_path)
+
+
+if __name__ == "__main__":
+    fire.Fire(main)

llama_toolchain/inference/quantization/scripts/run_quantize_checkpoint.sh

@@ -0,0 +1,31 @@
+#!/bin/bash
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+set -euo pipefail
+set -x
+
+cd $(git rev-parse --show-toplevel)
+
+MASTER_HOST=$1
+RUN_ID=$2
+CKPT_DIR=$3
+QUANT_CKPT_DIR=$4
+TOKENIZER_PATH=$5
+NNODES=$6
+NPROC=$7
+
+echo $MASTER_HOST, $RUN_ID, $CKPT_DIR, $QUANT_CKPT_DIR
+
+NCCL_NET=Socket NCCL_SOCKET_IFNAME=eth TIKTOKEN_CACHE_DIR="" \
+  torchrun \
+   --nnodes=$NNODES --nproc_per_node=$NPROC \
+   --rdzv_id=$RUN_ID \
+   --rdzv_conf='timeout=120' \
+   --rdzv_backend=c10d \
+   --rdzv_endpoint="${MASTER_HOST}:29502" \
+   quantize_checkpoint.py $CKPT_DIR $TOKENIZER_PATH $QUANT_CKPT_DIR

llama_toolchain/inference/quantization/test_fp8.py

@@ -0,0 +1,76 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# This software may be used and distributed in accordance with the terms of the Llama 3 Community License Agreement.
+
+import unittest
+
+import torch
+
+from fp8_impls import ffn_swiglu_fp8_dynamic, FfnQuantizeMode, quantize_fp8
+from hypothesis import given, settings, strategies as st
+from torch import Tensor
+
+
+@unittest.skipIf(
+    not torch.cuda.is_available()
+    or torch.cuda.get_device_properties(torch.cuda.current_device()).major < 9,
+    "Skip when H100 is not available",
+)
+class FP8Tests(unittest.TestCase):
+    @settings(deadline=None)
+    @given(
+        D=st.sampled_from([4096, 8192]),
+        HD_L=st.sampled_from([1280, 2560]),
+        B=st.sampled_from([1, 2]),
+        T=st.sampled_from([2048, 4096]),
+        UB=st.sampled_from([1000, 10000]),
+    )
+    def test_fp8_ffn(
+        self,
+        D: int,  # noqa
+        HD_L: int,
+        B: int,
+        T: int,
+        UB: float,
+    ) -> None:
+        x = torch.randn(size=(B, T, D), dtype=torch.bfloat16, device="cuda") * 0.1
+        w1 = torch.randn(size=(HD_L, D), dtype=torch.bfloat16, device="cuda") * 0.01
+        w3 = torch.randn(size=(HD_L, D), dtype=torch.bfloat16, device="cuda") * 0.01
+        w2 = torch.randn(size=(D, HD_L), dtype=torch.bfloat16, device="cuda") * 0.1
+
+        x_q = quantize_fp8(x, UB, mode=FfnQuantizeMode.FP8_ROWWISE)
+        w1_q = quantize_fp8(w1, UB, mode=FfnQuantizeMode.FP8_ROWWISE)
+        w3_q = quantize_fp8(w3, UB, mode=FfnQuantizeMode.FP8_ROWWISE)
+        w2_q = quantize_fp8(w2, UB, mode=FfnQuantizeMode.FP8_ROWWISE)
+
+        def ref_ffn(x: Tensor, w1: Tensor, w3: Tensor, w2: Tensor) -> Tensor:
+            (B, T, D) = x.shape  # noqa: N806
+            (HD_L, D_) = w1.shape  # noqa: N806
+            assert D_ == D
+
+            x1 = x.view(B * T, D) @ w1.T
+            x2 = x.view(B * T, D) @ w3.T
+
+            z = torch.nn.functional.silu(x1) * x2
+            return (z @ w2.T).view(B, T, D).to(torch.bfloat16)
+
+        v = ffn_swiglu_fp8_dynamic(x, w1_q, w3_q, w2_q)
+
+        # Fake quant
+        x = x_q.weight.bfloat16() * x_q.scale.unsqueeze(-1)
+        w1 = w1_q.weight.bfloat16() * w1_q.scale.unsqueeze(-1)
+        w3 = w3_q.weight.bfloat16() * w3_q.scale.unsqueeze(-1)
+        w2 = w2_q.weight.bfloat16() * w2_q.scale.unsqueeze(-1)
+
+        v_ref = ref_ffn(x, w1, w3, w2)
+
+        torch.testing.assert_close(v_ref, v, atol=4.0e-3, rtol=4.0e-3)
+
+
+if __name__ == "__main__":
+    unittest.main()

llama_toolchain/inference/server.py

@@ -0,0 +1,119 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+import asyncio
+import signal
+
+import fire
+
+from dotenv import load_dotenv
+
+from fastapi import FastAPI, HTTPException, Request
+from fastapi.responses import StreamingResponse
+
+from hydra_zen import instantiate
+
+from llama_toolchain.utils import get_default_config_dir, parse_config
+from .api.endpoints import ChatCompletionRequest, ChatCompletionResponseStreamChunk
+
+from .api_instance import get_inference_api_instance
+
+
+load_dotenv()
+
+
+GLOBAL_CONFIG = None
+
+
+def get_config():
+    return GLOBAL_CONFIG
+
+
+def handle_sigint(*args, **kwargs):
+    print("SIGINT or CTRL-C detected. Exiting gracefully", args)
+    loop = asyncio.get_event_loop()
+    for task in asyncio.all_tasks(loop):
+        task.cancel()
+    loop.stop()
+
+
+app = FastAPI()
+
+
+@app.on_event("startup")
+async def startup():
+    global InferenceApiInstance
+
+    config = get_config()
+
+    inference_config = instantiate(config["inference_config"])
+    InferenceApiInstance = await get_inference_api_instance(
+        inference_config,
+    )
+    await InferenceApiInstance.initialize()
+
+
+@app.on_event("shutdown")
+async def shutdown():
+    global InferenceApiInstance
+
+    print("shutting down")
+    await InferenceApiInstance.shutdown()
+
+
+# there's a single model parallel process running serving the model. for now,
+# we don't support multiple concurrent requests to this process.
+semaphore = asyncio.Semaphore(1)
+
+
+@app.post(
+    "/inference/chat_completion", response_model=ChatCompletionResponseStreamChunk
+)
+def chat_completion(request: Request, exec_request: ChatCompletionRequest):
+    if semaphore.locked():
+        raise HTTPException(
+            status_code=429,
+            detail="Only a single concurrent request allowed right now.",
+        )
+
+    async def sse_generator(event_gen):
+        try:
+            async for event in event_gen:
+                yield f"data: {event.json()}\n\n"
+                await asyncio.sleep(0.01)
+        except asyncio.CancelledError:
+            print("Generator cancelled")
+            await event_gen.aclose()
+        finally:
+            semaphore.release()
+
+    async def event_gen():
+        async for event in InferenceApiInstance.chat_completion(exec_request):
+            yield event
+
+    return StreamingResponse(
+        sse_generator(event_gen()),
+        media_type="text/event-stream",
+    )
+
+
+def main(config_path: str, port: int = 5000, disable_ipv6: bool = False):
+    global GLOBAL_CONFIG
+    config_dir = get_default_config_dir()
+    GLOBAL_CONFIG = parse_config(config_dir, config_path)
+
+    signal.signal(signal.SIGINT, handle_sigint)
+
+    import uvicorn
+
+    # FYI this does not do hot-reloads
+    listen_host = "::" if not disable_ipv6 else "0.0.0.0"
+    print(f"Listening on {listen_host}:{port}")
+    uvicorn.run(app, host=listen_host, port=port)
+
+
+if __name__ == "__main__":
+    fire.Fire(main)