fix

Author: rebel-eunji

.pre-commit-config.yaml

@@ -18,6 +18,7 @@ repos:
   rev: v1.7.7
   hooks:
   - id: actionlint
+    args: ["-config-file", "actionlint.yaml"]
 - repo: https://github.com/pre-commit/mirrors-clang-format
   rev: v21.1.2
   hooks:
@@ -68,11 +69,6 @@ repos:
     entry: tools/mypy.sh 1 "3.13"
     <<: *mypy_common
     stages: [manual] # Only run in CI
-  - id: shellcheck
-    name: Lint shell scripts
-    entry: tools/pre_commit/shellcheck.sh
-    language: script
-    types: [shell]
   - id: signoff-commit
     name: Sign-off Commit
     entry: bash

actionlint.yaml

@@ -0,0 +1,7 @@
+self-hosted-runner:
+  labels:
+    - rbln-sw-k8s-general
+    - rbln-sw-k8s-ca22-1
+    - rbln-sw-k8s-ca22-4
+    - fsw-r18s-atom-ext-rbln-sw-general-c1m12
+    - fsw-r18s-atom-ext-rbln-sw-general-c3m48a1

benchmarks/benchmark_serving.py

@@ -1125,32 +1125,37 @@ def main(args: argparse.Namespace):
         "--custom-output-len",
         type=int,
         default=256,
-        help="Number of output tokens per request, used only for custom dataset.",
+        help="Number of output tokens per request, "
+        "used only for custom dataset.",
     )
     custom_group.add_argument(
         "--custom-skip-chat-template",
         action="store_true",
-        help="Skip applying chat template to prompt, used only for custom dataset.",
+        help="Skip applying chat template to prompt, "
+        "used only for custom dataset.",
     )
 
     sonnet_group = parser.add_argument_group("sonnet dataset options")
     sonnet_group.add_argument(
         "--sonnet-input-len",
         type=int,
         default=550,
-        help="Number of input tokens per request, used only for sonnet dataset.",
+        help="Number of input tokens per request, "
+        "used only for sonnet dataset.",
     )
     sonnet_group.add_argument(
         "--sonnet-output-len",
         type=int,
         default=150,
-        help="Number of output tokens per request, used only for sonnet dataset.",
+        help="Number of output tokens per request, "
+        "used only for sonnet dataset.",
     )
     sonnet_group.add_argument(
         "--sonnet-prefix-len",
         type=int,
         default=200,
-        help="Number of prefix tokens per request, used only for sonnet dataset.",
+        help="Number of prefix tokens per request, "
+        "used only for sonnet dataset.",
     )
 
     sharegpt_group = parser.add_argument_group("sharegpt dataset options")
@@ -1167,13 +1172,15 @@ def main(args: argparse.Namespace):
         "--random-input-len",
         type=int,
         default=1024,
-        help="Number of input tokens per request, used only for random sampling.",
+        help="Number of input tokens per request, "
+        "used only for random sampling.",
     )
     random_group.add_argument(
         "--random-output-len",
         type=int,
         default=128,
-        help="Number of output tokens per request, used only for random sampling.",
+        help="Number of output tokens per request, "
+        "used only for random sampling.",
     )
     random_group.add_argument(
         "--random-range-ratio",

benchmarks/benchmark_serving_structured_output.py

@@ -147,7 +147,8 @@ def sample_requests(
                     f"__optional_field_{uuid.uuid4()}"
                 ] = {
                     "type": "string",
-                    "description": "An unique optional field to avoid cached schemas",
+                    "description": "An unique optional field "
+                    "to avoid cached schemas",
                 }
         else:
             json_schemas = [schema] * args.num_prompts

benchmarks/benchmark_utils.py

@@ -103,7 +103,8 @@ def write_to_json(filename: str, records: list) -> None:
             records,
             f,
             cls=InfEncoder,
-            default=lambda o: f"<{type(o).__name__} object is not JSON serializable>",
+            default=lambda o: f"<{type(o).__name__} "
+            "object is not JSON serializable>",
         )
 
 

examples/optimum/run_encoder_only.py

@@ -37,7 +37,8 @@ def compare_copy_prompt_task_result(scores: list[float], golden_json: str):
         golden_similarity = golden[i]["golden_similarity"]
         diff = abs(similarity - golden_similarity)
         print(
-            "Difference: {:.3f} Similarity : {:.3f}, Golden Similarity: {:.3f}".format(
+            "Difference: {:.3f} Similarity : {:.3f}, "
+            "Golden Similarity: {:.3f}".format(
                 diff, similarity, golden_similarity
             )
         )

examples/optimum/run_qwen3_reranker.py

@@ -35,7 +35,8 @@ def format_instruction(instruction, query, doc):
         },
         {
             "role": "user",
-            "content": f"<Instruct>: {instruction}\n\n<Query>: {query}\n\n<Document>: {doc}",
+            "content": f"<Instruct>: {instruction}\n\n"
+            "<Query>: {query}\n\n<Document>: {doc}",
         },
     ]
     return text

pyproject.toml

@@ -168,3 +168,17 @@ torchvision = { index = "pytorch-cpu" }
 name = "pytorch-cpu"
 url = "https://download.pytorch.org/whl/cpu"
 explicit = true
+
+[tool.typos.default]
+binary = false
+check-filename = false
+check-file = true
+unicode = true
+ignore-hex = true
+identifier-leading-digits = false
+locale = "en"
+extend-ignore-identifiers-re = ["NVML_*", ".*Unc.*", ".*_thw",
+    ".*UE8M0.*", ".*[UE4M3|ue4m3].*", ".*eles.*",
+     ".*[Tt]h[rR].*"]
+extend-ignore-words-re = []
+extend-ignore-re = []

tests/entrypoints/openai/test_basic.py

@@ -22,7 +22,7 @@
 import requests
 from utils import RemoteOpenAIServer
 
-MODEL_DIR = os.getenv("REBEL_VLLM_PRE_COMPILED_DIR")
+MODEL_DIR = os.getenv("REBEL_VLLM_PRE_COMPILED_DIR", ".")
 MODEL_NAME = MODEL_DIR + "/opt_125m_batch2"
 MAX_TOKENS = 1
 

tests/entrypoints/openai/test_chat.py

@@ -19,7 +19,7 @@
 import pytest_asyncio
 from utils import RemoteOpenAIServer
 
-MODEL_DIR = os.getenv("REBEL_VLLM_PRE_COMPILED_DIR")
+MODEL_DIR = os.getenv("REBEL_VLLM_PRE_COMPILED_DIR", ".")
 MODEL_NAME = MODEL_DIR + "/llama3_2-3b-128k_kv16k_batch4"
 MAX_TOKENS = 1
 
@@ -37,8 +37,7 @@ def monkeypatch_module():
 def server(request, monkeypatch_module):
     use_v1 = request.param
     monkeypatch_module.setenv("VLLM_USE_V1", "1" if use_v1 else "0")
-    args = []
-    with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
+    with RemoteOpenAIServer(MODEL_NAME, []) as remote_server:
         yield remote_server