fix: resolve CUTLASS fmin compatibility for DeepSeek-V4 init

Signed-off-by: khluu <khluu000@gmail.com>

.buildkite/scripts/install-kv-connectors.sh

@@ -0,0 +1,39 @@
+#!/bin/bash
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+set -euo pipefail
+
+REQUIREMENTS_FILE="${KV_CONNECTORS_REQUIREMENTS:-/vllm-workspace/requirements/kv_connectors.txt}"
+
+uv pip install --system -r "${REQUIREMENTS_FILE}"
+
+NIXL_METADATA=$(python3 - <<'PY'
+import importlib.metadata as metadata
+
+import torch
+
+cuda_version = torch.version.cuda
+if cuda_version is None:
+    raise SystemExit("torch.version.cuda is not set")
+
+print(cuda_version.split(".", 1)[0], metadata.version("nixl"))
+PY
+)
+read -r CUDA_MAJOR NIXL_VERSION <<<"${NIXL_METADATA}"
+
+# nixl>=1.1.0 can install multiple CUDA wheel variants. Keep only the variant
+# matching this CI image so nixl_ep_cpp links against the available libcudart.
+uv pip uninstall --system nixl-cu12 nixl-cu13 2>/dev/null || true
+uv pip install --system --no-deps "nixl-cu${CUDA_MAJOR}==${NIXL_VERSION}"
+
+python3 - <<'PY'
+import importlib.metadata as metadata
+
+for package_name in ("nixl", "nixl-cu12", "nixl-cu13"):
+    try:
+        version = metadata.version(package_name)
+    except metadata.PackageNotFoundError:
+        version = "not installed"
+    print(f"{package_name}: {version}")
+PY

.buildkite/test_areas/disaggregated.yaml

@@ -11,7 +11,7 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/nixl/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
 - label: Distributed FlashInfer NixlConnector PD accuracy (4 GPUs)
   key: distributed-flashinfer-nixlconnector-pd-accuracy-4-gpus
@@ -22,7 +22,7 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/nixl/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - FLASHINFER=1 bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
 
 - label: DP EP Distributed NixlConnector PD accuracy tests (4 GPUs)
@@ -34,7 +34,7 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/nixl/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - DP_EP=1 bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
 
 - label: CrossLayer KV layout Distributed NixlConnector PD accuracy tests (4 GPUs)
@@ -46,7 +46,7 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/nixl/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - CROSS_LAYERS_BLOCKS=True bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
 
 - label: Hybrid SSM NixlConnector PD accuracy tests (4 GPUs)
@@ -58,7 +58,7 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/nixl/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - HYBRID_SSM=1 bash v1/kv_connector/nixl_integration/config_sweep_accuracy_test.sh
 
 - label: MultiConnector (Nixl+Offloading) PD accuracy (2 GPUs)
@@ -73,7 +73,7 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/offloading/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - bash v1/kv_connector/nixl_integration/run_multi_connector_accuracy_test.sh
 
 - label: NixlConnector PD + Spec Decode acceptance (2 GPUs)
@@ -87,7 +87,7 @@ steps:
     - vllm/v1/worker/kv_connector_model_runner_mixin.py
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - bash v1/kv_connector/nixl_integration/config_sweep_spec_decode_test.sh
 
 - label: MultiConnector (Nixl+Offloading) PD edge cases (2 GPUs)
@@ -102,5 +102,5 @@ steps:
     - vllm/distributed/kv_transfer/kv_connector/v1/offloading/
     - tests/v1/kv_connector/nixl_integration/
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
-    - bash v1/kv_connector/nixl_integration/run_multi_connector_edge_case_test.sh
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
+    - bash v1/kv_connector/nixl_integration/run_multi_connector_edge_case_test.sh

.buildkite/test_areas/misc.yaml

@@ -86,7 +86,7 @@ steps:
     - tests/v1/metrics
     - tests/entrypoints/openai/correctness/test_lmeval.py
   commands:
-    - uv pip install --system -r /vllm-workspace/requirements/kv_connectors.txt
+    - bash /vllm-workspace/.buildkite/scripts/install-kv-connectors.sh
     - export VLLM_WORKER_MULTIPROC_METHOD=spawn
     # split the test to avoid interference
     - pytest -v -s -m 'not cpu_test' v1/core

tests/models/registry.py

@@ -335,7 +335,7 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
     "HYV3ForCausalLM": _HfExamplesInfo("tencent/Hy3-preview", trust_remote_code=True),
     "HyperCLOVAXForCausalLM": _HfExamplesInfo(
         "naver-hyperclovax/HyperCLOVAX-SEED-Think-14B",
-        trust_remote_code=True,
+        min_transformers_version="5.9.0",
     ),
     "InternLMForCausalLM": _HfExamplesInfo(
         "internlm/internlm-chat-7b", trust_remote_code=True

vllm/models/deepseek_v4/nvidia/ops/sparse_attn_compress_cutedsl.py

@@ -320,11 +320,11 @@ class SparseAttnCompressNormRopeStoreC4Kernel:
                 bits = _recast_val(scale_raw, Uint32)
                 ue8m0 = ((bits + Uint32(0x7FFFFF)) >> Uint32(23)) & Uint32(0xFF)
                 inv_scale = _recast_val((Uint32(254) - ue8m0) << Uint32(23), Float32)
-                y0 = cute.arch.fmin(
+                y0 = cutlass.min(
                     cute.arch.fmax(q0 * inv_scale, Float32(-self.fp8_max)),
                     Float32(self.fp8_max),
                 )
-                y1 = cute.arch.fmin(
+                y1 = cutlass.min(
                     cute.arch.fmax(q1 * inv_scale, Float32(-self.fp8_max)),
                     Float32(self.fp8_max),
                 )
@@ -978,11 +978,11 @@ class SparseAttnNormRopeStoreKernel:
                 bits = _recast_val(scale_raw, Uint32)
                 ue8m0 = ((bits + Uint32(0x7FFFFF)) >> Uint32(23)) & Uint32(0xFF)
                 inv_scale = _recast_val((Uint32(254) - ue8m0) << Uint32(23), Float32)
-                y0 = cute.arch.fmin(
+                y0 = cutlass.min(
                     cute.arch.fmax(q0 * inv_scale, Float32(-self.fp8_max)),
                     Float32(self.fp8_max),
                 )
-                y1 = cute.arch.fmin(
+                y1 = cutlass.min(
                     cute.arch.fmax(q1 * inv_scale, Float32(-self.fp8_max)),
                     Float32(self.fp8_max),
                 )

vllm/transformers_utils/config.py

@@ -101,6 +101,7 @@ _CONFIG_REGISTRY: dict[str, type[PretrainedConfig]] = LazyConfigDict(
     fireredlid="FireRedLIDConfig",
     funaudiochat="FunAudioChatConfig",
     granite4_vision="Granite4VisionConfig",
+    hyperclovax="HyperCLOVAXConfig",
     hyperclovax_vlm="HCXVisionConfig",
     hunyuan_vl="HunYuanVLConfig",
     hy_v3="HYV3Config",
@@ -430,9 +431,9 @@ def patch_legacy_rope_type(rope_parameters: dict[str, Any] | None) -> None:
         if "rope_type" not in rope_parameters and "type" in rope_parameters:
             rope_parameters["rope_type"] = rope_parameters["type"]
             logger.info("Replacing legacy 'type' key with 'rope_type'")
-        # Case 3: No rope_type field at all - cannot determine RoPE type, raise error
+        # Case 3: No rope_type field present - nothing to patch
         if "rope_type" not in rope_parameters:
-            raise ValueError("rope_parameters should have a 'rope_type' key")
+            return
         # Patch legacy rope_type values with warning
         if rope_parameters["rope_type"] == "su":
             rope_parameters["rope_type"] = "longrope"