Files
MOVZXandHinata 2066d3143f [Mini-VLLM] Cut non-OpenAI surfaces, dead kernels, Mistral stack
- Remove Anthropic, Cohere, SageMaker, decisions, structured-decisions,
  rl, render-launcher and batch entrypoints; keep scale-out token routes
- Drop the Swagger static bundle and the /docs, /redoc routes
- Delete device-keyed tuning configs for fused_moe, quant utils, mamba
  select-state and helion; no sm86 entry ever existed for any of them
- Remove ten unused rope variants plus the fbgemm_fp8 and fp_quant
  quantizers and their config, metrics and loader plumbing
- Cut the CPU device build: csrc/cpu, cpu_extension.cmake, cpu
  requirements and the setup.py cpu target path
- Delete 29 environment variables nothing reads and 14 chat-template
  fallbacks for models this fork no longer registers
- Remove nine dead CUDA ops (deepseek_v4, kimi_k3, minimax, muse_spark,
  dsv3 gemm) with their kernels, bindings, wrappers and CMake entries
- Delete the connector test suites that targeted removed connectors
- Cut the Mistral tokenizer and parser stack, its tests, and the
  mistral_common and ijson dependency pins
- Restore the Qwen4Exp registry rows and nvidia/mtp.py that an earlier
  cut deleted while keeping the model tree

Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
2026-10-19 17:10:56 +07:00

81 lines
2.4 KiB
Python

# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from dataclasses import dataclass, field
from typing import TYPE_CHECKING
from vllm.logger import init_logger
from vllm.tokenizers import TokenizerLike
from vllm.tokenizers.registry import (
cached_tokenizer_from_config,
tokenizer_args_from_config,
)
from vllm.utils.import_utils import resolve_obj_by_qualname
from .base import BaseRenderer
if TYPE_CHECKING:
from vllm.config import VllmConfig
logger = init_logger(__name__)
_VLLM_RENDERERS = {
"hf": ("hf", "HfRenderer"),
}
@dataclass
class RendererRegistry:
# Renderer mode -> (renderer module, renderer class)
renderers: dict[str, tuple[str, str]] = field(default_factory=dict)
def register(self, renderer_mode: str, module: str, class_name: str) -> None:
if renderer_mode in self.renderers:
logger.warning(
"%s.%s is already registered for renderer_mode=%r. "
"It is overwritten by the new one.",
module,
class_name,
renderer_mode,
)
self.renderers[renderer_mode] = (module, class_name)
return None
def load_renderer_cls(self, renderer_mode: str) -> type[BaseRenderer]:
if renderer_mode not in self.renderers:
raise ValueError(f"No renderer registered for {renderer_mode=!r}.")
module, class_name = self.renderers[renderer_mode]
logger.debug_once(f"Loading {class_name} for {renderer_mode=!r}")
return resolve_obj_by_qualname(f"{module}.{class_name}")
def load_renderer(
self,
renderer_mode: str,
config: "VllmConfig",
tokenizer: TokenizerLike | None,
) -> BaseRenderer:
renderer_cls = self.load_renderer_cls(renderer_mode)
return renderer_cls(config, tokenizer)
RENDERER_REGISTRY = RendererRegistry(
{
mode: (f"vllm.renderers.{mod_relname}", cls_name)
for mode, (mod_relname, cls_name) in _VLLM_RENDERERS.items()
}
)
"""The global `RendererRegistry` instance."""
def renderer_from_config(config: "VllmConfig", **kwargs):
model_config = config.model_config
tokenizer = cached_tokenizer_from_config(model_config, **kwargs)
renderer_mode, *_ = tokenizer_args_from_config(model_config, **kwargs)
return RENDERER_REGISTRY.load_renderer(renderer_mode, config, tokenizer)