[Model] New model support for Motif-1-Tiny (#23414)

Signed-off-by: ca1207 <ca1207zzz@gmail.com>
Signed-off-by: TaehyunKim <73943231+ca1207@users.noreply.github.com>
Co-authored-by: WyldeCat <skan1543@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
TaehyunKim
2025-09-10 23:29:40 -07:00
committed by GitHub
co-authored by WyldeCat Jee Jee Li
parent e2b1f863aa
commit 9bd831f501
13 changed files with 871 additions and 4 deletions
+3
View File
@@ -291,6 +291,9 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
"MistralForCausalLM": _HfExamplesInfo("mistralai/Mistral-7B-Instruct-v0.1"),
"MixtralForCausalLM": _HfExamplesInfo("mistralai/Mixtral-8x7B-Instruct-v0.1", # noqa: E501
{"tiny": "TitanML/tiny-mixtral"}), # noqa: E501
"MotifForCausalLM": _HfExamplesInfo("Motif-Technologies/Motif-2.6B",
trust_remote_code=True,
v0_only=True),
"MptForCausalLM": _HfExamplesInfo("mpt", is_available_online=False),
"MPTForCausalLM": _HfExamplesInfo("mosaicml/mpt-7b"),
"NemotronForCausalLM": _HfExamplesInfo("nvidia/Minitron-8B-Base"),
+3 -2
View File
@@ -63,8 +63,9 @@ def can_initialize(model_arch: str, monkeypatch: pytest.MonkeyPatch,
_initialize_kv_caches_v1), monkeypatch.context() as m):
if model_info.v0_only:
m.setenv("VLLM_USE_V1", "0")
if model_arch == "Phi4FlashForCausalLM":
# Phi4FlashForCausalLM only supports DIFFERENTIAL_FLASH_ATTN backend
if model_arch in ("Phi4FlashForCausalLM", "MotifForCausalLM"):
# Phi4FlashForCausalLM and MotifForCausalLM
# only supports DIFFERENTIAL_FLASH_ATTN backend
m.setenv("VLLM_ATTENTION_BACKEND", "DIFFERENTIAL_FLASH_ATTN")
if model_arch == "GptOssForCausalLM":
# FIXME: A hack to bypass FA3 assertion because our CI's L4 GPU