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 14, 05:30 UTC
[CI] [Hybrid] Bump min transformers version for Bamba and Jamba (#22908)
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
Thomas Parnell committed
Aug 14, 2025 at 20:01 UTC
ab9f2cfd1942f7ddfee658ce86ea96b4789862af
1 file changed
+2
-2
tests/models/registry.py
+2
-2
@@ -151,7 +151,7 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
151
"BailingMoeForCausalLM": _HfExamplesInfo("inclusionAI/Ling-lite-1.5",
152
trust_remote_code=True),
153
"BambaForCausalLM": _HfExamplesInfo("ibm-ai-platform/Bamba-9B-v1",
154
- min_transformers_version="4.55.1",
154
+ min_transformers_version="4.56.0",
155
extras={"tiny": "hmellor/tiny-random-BambaForCausalLM"}), # noqa: E501
156
"BloomForCausalLM": _HfExamplesInfo("bigscience/bloom-560m",
157
{"1b": "bigscience/bloomz-1b1"}),
@@ -227,7 +227,7 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
227
trust_remote_code=True),
228
"JAISLMHeadModel": _HfExamplesInfo("inceptionai/jais-13b-chat"),
229
"JambaForCausalLM": _HfExamplesInfo("ai21labs/AI21-Jamba-1.5-Mini",
230
- min_transformers_version="4.55.1",
230
+ min_transformers_version="4.56.0",
231
extras={
232
"tiny": "ai21labs/Jamba-tiny-dev",
233
"random": "ai21labs/Jamba-tiny-random", # noqa: E501