[Mini-vLLM] SM86-only cut: models, backends, Rust, tests

- Keep only Qwen3.5, Qwen3.5-MTP and Qwen4Exp model paths; trim the model registry to 41 entries
- Delete idle attention backends (MLA, FlashInfer, ROCm-only, quant backends) and dead GDN CuTeDSL kernels
- Remove inc/quark quant loaders; keep compressed-tensors and Marlin
- Drop the Rust frontend, pooling entrypoints, and idle test suites
- Rebuild kernels for sm_80;sm_86 and strip the extensions
- Verified: full import sweep clean (optional deps only) and a dummy-load Qwen3.5 engine run on RTX 3090 via the Triton GDN path

Co-authored-by: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
This commit is contained in:
2026-10-18 19:19:56 +07:00
co-authored by Hinata
parent c5ab42bc0f
commit e61dcb9143
2728 changed files with 159 additions and 950401 deletions
@@ -41,12 +41,7 @@ from vllm.config.utils import getattr_iter
from vllm.distributed import get_pp_group, get_tp_group
from vllm.distributed.utils import get_pp_indices
from vllm.logger import init_logger
from vllm.model_executor.layers.attention import (
Attention,
EncoderOnlyAttention,
MLAAttention,
)
from vllm.model_executor.layers.attention.mla_attention import get_mla_dims
from vllm.model_executor.layers.attention import Attention, EncoderOnlyAttention
from vllm.model_executor.layers.attention_layer_base import AttentionLayerBase
from vllm.model_executor.layers.fused_moe import MoERunner
from vllm.model_executor.models.interfaces import (
@@ -91,6 +86,11 @@ if TYPE_CHECKING:
from vllm.config import VllmConfig
try:
from vllm.model_executor.layers.attention.mla_attention import MLAAttention
except ImportError: # MLA support removed from this build
MLAAttention = type("AbsentMLAAttention", (), {})
logger = init_logger(__name__)
VLLM_ATTN_ATTR = "attn"
@@ -657,6 +657,10 @@ class Base(
f"Layer {i} ({attn_prefix}) did not fuse as MLA, but the rest "
"of the model did, so its latent KV cache cannot be allocated."
)
from vllm.model_executor.layers.attention.mla_attention import (
get_mla_dims,
)
dims = get_mla_dims(self.model_config)
kwargs.update(
qk_nope_head_dim=dims.qk_nope_head_dim,