[Mini-vLLM] Cut dead subsystems: models, entrypoints, backends, utils, tests

- Drop dead model modules, keeping files live imports require
- Remove anthropic, cohere, codegen entrypoints and batch STT modes
- Delete unused attention backends, MoE experts, and NPU/XPU util leaves
- Strip dead third_party trees and ROCm leaves with no CUDA path
- Remove mooncake EC connectors: dependency missing since branch base
- Re-audit verified by full import sweep and Qwen3.5 smoke test

Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
This commit is contained in:
2026-10-19 00:20:48 +07:00
co-authored by Hinata
parent 6a95f62e2f
commit 81074e94a3
940 changed files with 11 additions and 380533 deletions
@@ -1,63 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
import pytest
from vllm import SamplingParams
from vllm.entrypoints.common.offline import OfflineInferenceMixin
from vllm.exceptions import VLLMValidationError
from vllm.lora.request import LoRARequest
@pytest.fixture
def mixin() -> OfflineInferenceMixin:
return object.__new__(OfflineInferenceMixin)
def test_rejects_mismatched_params(mixin: OfflineInferenceMixin):
with pytest.raises(VLLMValidationError) as exc_info:
mixin._params_to_seq([SamplingParams()], num_requests=2)
assert str(exc_info.value) == (
"The lengths of prompts (2) and params (1) must be the same."
)
assert exc_info.value.parameter is None
assert exc_info.value.value is None
def test_rejects_mismatched_lora_requests(mixin: OfflineInferenceMixin):
with pytest.raises(VLLMValidationError) as exc_info:
mixin._lora_request_to_seq([None], num_requests=2)
assert str(exc_info.value) == (
"The lengths of prompts (2) and lora_request (1) must be the same."
)
assert exc_info.value.parameter is None
assert exc_info.value.value is None
def test_rejects_mismatched_priority(mixin: OfflineInferenceMixin):
with pytest.raises(VLLMValidationError) as exc_info:
mixin._priority_to_seq([0], num_requests=2)
assert str(exc_info.value) == (
"The lengths of prompts (2) and priority (1) must be the same."
)
assert exc_info.value.parameter is None
assert exc_info.value.value is None
def test_matching_lengths_pass_through(mixin: OfflineInferenceMixin):
lora = LoRARequest("l", 1, "/tmp/l")
assert mixin._params_to_seq([SamplingParams()], num_requests=1) == [
SamplingParams()
]
assert mixin._lora_request_to_seq([lora], num_requests=1) == [lora]
assert mixin._priority_to_seq([3], num_requests=1) == [3]
def test_scalar_priority_broadcast(mixin: OfflineInferenceMixin):
assert mixin._priority_to_seq(5, num_requests=1) == [5]
assert mixin._priority_to_seq(5, num_requests=3) == [5, 5, 5]
assert mixin._priority_to_seq(None, num_requests=3) == [0, 0, 0]