[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>
This commit is contained in:
2026-10-19 17:10:56 +07:00
co-authored by Hinata
parent bd70a61c20
commit 2066d3143f
1086 changed files with 122 additions and 291919 deletions
@@ -13,31 +13,7 @@ CHAT_TEMPLATES_DIR = Path(__file__).parent
ChatTemplatePath: TypeAlias = Path | Callable[[str], Path | None]
def _get_minicpmv_chat_template_fallback(tokenizer_name_or_path: str) -> Path | None:
# MiniCPM-V-4.5 version uses a dedicated template
if "4.5" in tokenizer_name_or_path or "4_5" in tokenizer_name_or_path:
return CHAT_TEMPLATES_DIR / "template_minicpmv45.jinja"
# Other versions use chatml template
return CHAT_TEMPLATES_DIR / "template_chatml.jinja"
_MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK: dict[str, ChatTemplatePath] = {
"blip-2": CHAT_TEMPLATES_DIR / "template_blip2.jinja",
"clip": CHAT_TEMPLATES_DIR / "template_basic.jinja",
"colpali": CHAT_TEMPLATES_DIR / "template_basic.jinja",
"deepseek_ocr": CHAT_TEMPLATES_DIR / "template_deepseek_ocr.jinja",
"deepseek_ocr2": CHAT_TEMPLATES_DIR / "template_deepseek_ocr.jinja",
"unlimited-ocr": CHAT_TEMPLATES_DIR / "template_deepseek_ocr.jinja",
"unlimited_ocr": CHAT_TEMPLATES_DIR / "template_deepseek_ocr.jinja",
"deepseek_vl_v2": CHAT_TEMPLATES_DIR / "template_deepseek_vl2.jinja",
"minicpmv": _get_minicpmv_chat_template_fallback,
"minicpmv4_6": _get_minicpmv_chat_template_fallback,
"minicpmv4_7": _get_minicpmv_chat_template_fallback,
"paligemma": CHAT_TEMPLATES_DIR / "template_basic.jinja",
"siglip": CHAT_TEMPLATES_DIR / "template_basic.jinja",
"siglip2": CHAT_TEMPLATES_DIR / "template_basic.jinja",
}
_MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK: dict[str, ChatTemplatePath] = {}
def register_chat_template_fallback_path(
@@ -1,3 +0,0 @@
{%- for message in messages -%}
{{- message['content'] -}}
{%- endfor -%}
@@ -1,11 +0,0 @@
{%- for message in messages -%}
{%- if message['role'] == 'user' -%}
{{- 'Question: ' + message['content'] + ' ' -}}
{%- elif message['role'] == 'assistant' -%}
{{- 'Answer: ' + message['content'] + ' ' -}}
{%- endif -%}
{%- endfor -%}
{%- if add_generation_prompt -%}
{{- 'Answer:' -}}
{% endif %}
@@ -1,10 +0,0 @@
{%- for message in messages -%}
{{- '<|im_start|>' + message['role'] + '\n' + message['content'] -}}
{%- if (loop.last and add_generation_prompt) or not loop.last -%}
{{- '<|im_end|>' + '\n' -}}
{%- endif -%}
{%- endfor -%}
{%- if add_generation_prompt and messages[-1]['role'] != 'assistant' -%}
{{- '<|im_start|>assistant\n' -}}
{%- endif -%}
@@ -1,14 +0,0 @@
{%- if messages[0]['role'] == 'system' -%}
{%- set system_message = messages[0]['content'] -%}
{%- set messages = messages[1:] -%}
{%- else -%}
{% set system_message = '' -%}
{%- endif -%}
{{ bos_token + system_message }}
{%- for message in messages -%}
{%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}
{%- endif -%}
{{ message['content'] }}
{%- endfor -%}
@@ -1,23 +0,0 @@
{%- if messages[0]['role'] == 'system' -%}
{%- set system_message = messages[0]['content'] -%}
{%- set messages = messages[1:] -%}
{%- else -%}
{% set system_message = '' -%}
{%- endif -%}
{{ bos_token + system_message }}
{%- for message in messages -%}
{%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}
{%- endif -%}
{%- if message['role'] == 'user' -%}
{{ '<|User|>: ' + message['content'] + '\n\n' }}
{%- elif message['role'] == 'assistant' -%}
{{ '<|Assistant|>: ' + message['content'] + eos_token + '\n\n' }}
{%- endif -%}
{%- endfor -%}
{%- if add_generation_prompt -%}
{{ '<|Assistant|>: ' }}
{%- endif -%}
@@ -1,13 +0,0 @@
{% set messages = conversations[0] if conversations else [] -%}
{% if messages and messages[0]['role'] == 'system' -%}
{% set loop_messages = messages[1:] -%}
{% else -%}
{% set loop_messages = messages -%}
{% endif -%}
{% for message in loop_messages -%}
{% if message['role'] == 'user' -%}
<|im_kimia_user_msg_start|>{{ message['content'] }}<|im_msg_end|><|im_kimia_assistant_msg_start|>
{%- elif message['role'] == 'assistant' -%}
{{ message['content'] }}<|im_kimia_text_eos|>
{%- endif -%}
{% endfor -%}
@@ -1,93 +0,0 @@
{%- set enable_thinking = enable_thinking | default(false) %}
{%- if tools %}
{{- '<|im_start|>system\n' }}
{%- if messages[0].role == 'system' %}
{{- messages[0].content + '\n\n' }}
{%- endif %}
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
{%- for tool in tools %}
{{- "\n" }}
{{- tool | tojson }}
{%- endfor %}
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
{%- else %}
{%- if messages[0].role == 'system' %}
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
{%- for message in messages[::-1] %}
{%- set index = (messages|length - 1) - loop.index0 %}
{%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
{%- set ns.multi_step_tool = false %}
{%- set ns.last_query_index = index %}
{%- endif %}
{%- endfor %}
{%- for message in messages %}
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
{%- elif message.role == "assistant" %}
{%- set content = message.content %}
{%- set reasoning = '' %}
{%- if message.reasoning is defined and message.reasoning is not none %}
{%- set reasoning = message.reasoning %}
{%- else %}
{%- if '</think>' in message.content %}
{%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
{%- set reasoning = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
{%- endif %}
{%- endif %}
{%- if loop.index0 > ns.last_query_index %}
{%- if loop.last or (not loop.last and reasoning) %}
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
{%- else %}
{{- '<|im_start|>' + message.role + '\n' + content }}
{%- endif %}
{%- else %}
{{- '<|im_start|>' + message.role + '\n' + content }}
{%- endif %}
{%- if message.tool_calls %}
{%- for tool_call in message.tool_calls %}
{%- if (loop.first and content) or (not loop.first) %}
{{- '\n' }}
{%- endif %}
{%- if tool_call.function %}
{%- set tool_call = tool_call.function %}
{%- endif %}
{{- '<tool_call>\n{"name": "' }}
{{- tool_call.name }}
{{- '", "arguments": ' }}
{%- if tool_call.arguments is string %}
{{- tool_call.arguments }}
{%- else %}
{{- tool_call.arguments | tojson }}
{%- endif %}
{{- '}\n</tool_call>' }}
{%- endfor %}
{%- endif %}
{{- '<|im_end|>\n' }}
{%- elif message.role == "tool" %}
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
{{- '<|im_start|>user' }}
{%- endif %}
{{- '\n<tool_response>\n' }}
{{- message.content }}
{{- '\n</tool_response>' }}
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
{{- '<|im_end|>\n' }}
{%- endif %}
{%- endif %}
{%- endfor %}
{%- if add_generation_prompt %}
{{- '<|im_start|>assistant\n' }}
{%- if enable_thinking is defined and enable_thinking is false %}
{{- '<think>\n\n</think>\n\n' }}
{%- endif %}
{%- if enable_thinking is defined and enable_thinking is true %}
{{- '<think>\n' }}
{%- endif %}
{%- endif %}
@@ -1,65 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from transformers import PreTrainedConfig
class Mimo_VLVisionConfig(PreTrainedConfig):
model_type = "mimovl"
base_config_key = "vision_config"
def __init__(
self,
depth=28,
hidden_size=1280,
hidden_act="silu",
intermediate_size=4608,
num_heads=32,
in_channels=3,
patch_size=16,
spatial_merge_size=2,
temporal_patch_size=2,
tokens_per_second=2,
window_size=128,
out_hidden_size=2048,
fullatt_block_indexes=None,
initializer_range=0.02,
kv_channels=64, # HACK
qk_channels=64,
num_query_groups=4,
num_key_value_heads=8,
vit_window_attn_types=None,
visual_token_window_size=64,
**kwargs,
):
super().__init__(**kwargs)
self.depth = depth
self.hidden_size = hidden_size
self.hidden_act = hidden_act
self.intermediate_size = intermediate_size
self.num_heads = num_heads
# Support GQA: if num_key_value_heads is not provided,
# default to num_heads (MHA)
if num_key_value_heads is None:
num_key_value_heads = num_heads
self.num_key_value_heads = num_key_value_heads
self.in_channels = in_channels
self.patch_size = patch_size
self.spatial_merge_size = spatial_merge_size
self.temporal_patch_size = temporal_patch_size
self.tokens_per_second = tokens_per_second
self.window_size = window_size
self.fullatt_block_indexes = (
fullatt_block_indexes
if fullatt_block_indexes is not None
else [7, 15, 23, 31]
)
self.out_hidden_size = out_hidden_size
self.initializer_range = initializer_range
self.kv_channels = kv_channels
self.qk_channels = qk_channels
self.num_query_groups = num_query_groups
self.vit_window_attn_types = vit_window_attn_types or [-1] * depth
self.visual_token_window_size = visual_token_window_size
+1 -26
View File
@@ -21,7 +21,6 @@ from transformers.video_processing_utils import BaseVideoProcessor
from typing_extensions import TypeVar
from vllm.logger import init_logger
from vllm.transformers_utils import processors
from vllm.transformers_utils.repo_utils import (
get_hf_file_to_dict,
get_non_weight_snapshot_path,
@@ -151,18 +150,6 @@ def get_processor_config(
return {}
def get_processor_cls_name_from_config(
processor_name: str,
revision: str | None = "main",
) -> str | None:
config = get_processor_config(processor_name, revision=revision)
if not config:
return None
return config["processor_class"]
def get_video_processor_cls_name_from_config(
processor_name: str,
revision: str | None = "main",
@@ -223,20 +210,8 @@ def get_processor(
processor_name = get_non_weight_snapshot_path(
processor_name, revision=revision, token=kwargs.get("token")
)
registered_cls_name = get_processor_cls_name_from_config(
processor_name, revision=revision
)
registered_processor_cls = (
getattr(processors, registered_cls_name, None)
if registered_cls_name
else None
)
registered_processor_cls = cast(type[_P] | None, registered_processor_cls)
# Use registered processor class when it's available
# and explicit processor_cls is not set.
if isinstance(processor_cls, tuple) or processor_cls == ProcessorMixin:
_processor_cls = registered_processor_cls or AutoProcessor
processor = _processor_cls.from_pretrained(
processor = AutoProcessor.from_pretrained(
processor_name,
*args,
revision=revision,
@@ -1,57 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
"""Multi-modal processors may be defined in this directory for the following
reasons:
- There is no processing file defined by HF Hub or Transformers library.
- There is a need to override the existing processor to support vLLM.
"""
import importlib
__all__ = [
"BagelProcessor",
"CohereASRProcessor",
"Cosmos3EdgeProcessor",
"DeepseekVLV2Processor",
"FireRedASR2Processor",
"FunASRProcessor",
"GLM4VProcessor",
"H2OVLProcessor",
"Moondream3Processor",
"IsaacProcessor",
"KimiAudioProcessor",
"KimiK25Processor",
"KimiK3Processor",
"MiMoOmniProcessor",
"MiniCPMOProcessor",
"MiniCPMVProcessor",
"MiniMaxM3VLImageProcessor",
"MiniMaxM3VLVideoProcessor",
"MiniMaxVLProcessor",
"MistralCommonPixtralProcessor",
"MistralCommonVoxtralProcessor",
"NanoNemotronVLProcessor",
"NemotronVLProcessor",
"LlamaNemotronVLEmbedProcessor",
"NVLMProcessor",
"OpenVLAProcessor",
"OvisProcessor",
"Ovis2_5Processor",
"Step3VLProcessor",
]
_CLASS_TO_MODULE: dict[str, str] = {}
def __getattr__(name: str):
if name in _CLASS_TO_MODULE:
module_name = _CLASS_TO_MODULE[name]
module = importlib.import_module(module_name)
return getattr(module, name)
raise AttributeError(f"module 'processors' has no attribute '{name}'")
def __dir__():
return sorted(list(__all__))