A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 04:45 UTC
[CI Fix] Pin tests/models/registry.py MiniMaxText01ForCausalLM to revision due to model changes (#20199)
Signed-off-by: mgoin <mgoin64@gmail.com>
Michael Goin committed
Jun 28, 2025 at 14:43 UTC
c329ceca6dd7263a65c7913a14de943266a38088
2 files changed
+9
-1
tests/models/registry.py
+8
-1
@@ -70,6 +70,12 @@ class _HfExamplesInfo:
70
length that is too large to fit into memory in CI.
71
"""
72
73
+ revision: Optional[str] = None
74
+ """
75
+ The specific revision (commit hash, tag, or branch) to use for the model.
76
+ If not specified, the default revision will be used.
77
+ """
78
+
79
def check_transformers_version(
80
self,
81
*,
@@ -207,7 +213,8 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
213
"MiniCPM3ForCausalLM": _HfExamplesInfo("openbmb/MiniCPM3-4B",
214
trust_remote_code=True),
215
"MiniMaxText01ForCausalLM": _HfExamplesInfo("MiniMaxAI/MiniMax-Text-01",
210
- trust_remote_code=True),
216
+ trust_remote_code=True,
217
+ revision="a59aa9cbc53b9fb8742ca4e9e1531b9802b6fdc3"), # noqa: E501
218
"MiniMaxM1ForCausalLM": _HfExamplesInfo("MiniMaxAI/MiniMax-M1-40k",
219
trust_remote_code=True),
220
"MistralForCausalLM": _HfExamplesInfo("mistralai/Mistral-7B-Instruct-v0.1"),
tests/models/test_initialization.py
+1
@@ -88,6 +88,7 @@ def test_can_initialize(model_arch: str, monkeypatch: pytest.MonkeyPatch):
88
model_info.default,
89
tokenizer=model_info.tokenizer,
90
tokenizer_mode=model_info.tokenizer_mode,
91
+ revision=model_info.revision,
92
speculative_config={
93
"model": model_info.speculative_model,
94
"num_speculative_tokens": 1,