[Misc] Modify the organization of GLM series (#22171)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
@@ -53,7 +53,7 @@ AITER_MODEL_LIST = [
|
||||
marks=[pytest.mark.core_model, pytest.mark.cpu_model],
|
||||
),
|
||||
pytest.param(
|
||||
"THUDM/chatglm3-6b", # chatglm (text-only)
|
||||
"zai-org/chatglm3-6b", # chatglm (text-only)
|
||||
),
|
||||
pytest.param(
|
||||
"meta-llama/Llama-3.2-1B-Instruct", # llama
|
||||
|
||||
@@ -355,7 +355,7 @@ VLM_TEST_SETTINGS = {
|
||||
num_logprobs=10,
|
||||
),
|
||||
"glm4v": VLMTestInfo(
|
||||
models=["THUDM/glm-4v-9b"],
|
||||
models=["zai-org/glm-4v-9b"],
|
||||
test_type=VLMTestType.IMAGE,
|
||||
prompt_formatter=lambda img_prompt: f"<|user|>\n{img_prompt}<|assistant|>", # noqa: E501
|
||||
single_image_prompts=IMAGE_ASSETS.prompts({
|
||||
@@ -374,7 +374,7 @@ VLM_TEST_SETTINGS = {
|
||||
marks=[large_gpu_mark(min_gb=32)],
|
||||
),
|
||||
"glm4_1v": VLMTestInfo(
|
||||
models=["THUDM/GLM-4.1V-9B-Thinking"],
|
||||
models=["zai-org/GLM-4.1V-9B-Thinking"],
|
||||
test_type=(VLMTestType.IMAGE, VLMTestType.MULTI_IMAGE),
|
||||
prompt_formatter=lambda img_prompt: f"<|user|>\n{img_prompt}<|assistant|>", # noqa: E501
|
||||
img_idx_to_prompt=lambda idx: "<|begin_of_image|><|image|><|end_of_image|>", # noqa: E501
|
||||
@@ -388,7 +388,7 @@ VLM_TEST_SETTINGS = {
|
||||
marks=[large_gpu_mark(min_gb=32)],
|
||||
),
|
||||
"glm4_1v-video": VLMTestInfo(
|
||||
models=["THUDM/GLM-4.1V-9B-Thinking"],
|
||||
models=["zai-org/GLM-4.1V-9B-Thinking"],
|
||||
# GLM4.1V require include video metadata for input
|
||||
test_type=VLMTestType.CUSTOM_INPUTS,
|
||||
max_model_len=4096,
|
||||
|
||||
@@ -271,8 +271,8 @@ def _test_processing_correctness_one(
|
||||
"microsoft/Florence-2-base",
|
||||
"adept/fuyu-8b",
|
||||
"google/gemma-3-4b-it",
|
||||
"THUDM/glm-4v-9b",
|
||||
"THUDM/GLM-4.1V-9B-Thinking",
|
||||
"zai-org/glm-4v-9b",
|
||||
"zai-org/GLM-4.1V-9B-Thinking",
|
||||
"ibm-granite/granite-speech-3.3-2b",
|
||||
"h2oai/h2ovl-mississippi-800m",
|
||||
"internlm/Intern-S1",
|
||||
|
||||
@@ -9,7 +9,7 @@ from vllm.multimodal import MULTIMODAL_REGISTRY
|
||||
from ...utils import build_model_context
|
||||
|
||||
|
||||
@pytest.mark.parametrize("model_id", ["THUDM/GLM-4.1V-9B-Thinking"])
|
||||
@pytest.mark.parametrize("model_id", ["zai-org/GLM-4.1V-9B-Thinking"])
|
||||
@pytest.mark.parametrize("expected_toks_per_frame", [299])
|
||||
@pytest.mark.parametrize("num_frames", [32, 128])
|
||||
@pytest.mark.parametrize("fps, expected_grid_t", [(1, 5), (2, 10)])
|
||||
|
||||
@@ -153,7 +153,7 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
|
||||
extras={"tiny": "hmellor/tiny-random-BambaForCausalLM"}), # noqa: E501
|
||||
"BloomForCausalLM": _HfExamplesInfo("bigscience/bloom-560m",
|
||||
{"1b": "bigscience/bloomz-1b1"}),
|
||||
"ChatGLMModel": _HfExamplesInfo("THUDM/chatglm3-6b",
|
||||
"ChatGLMModel": _HfExamplesInfo("zai-org/chatglm3-6b",
|
||||
trust_remote_code=True,
|
||||
max_transformers_version="4.48"),
|
||||
"ChatGLMForConditionalGeneration": _HfExamplesInfo("thu-coai/ShieldLM-6B-chatglm3", # noqa: E501
|
||||
@@ -187,8 +187,8 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
|
||||
"Gemma3ForCausalLM": _HfExamplesInfo("google/gemma-3-1b-it"),
|
||||
"Gemma3nForConditionalGeneration": _HfExamplesInfo("google/gemma-3n-E2B-it", # noqa: E501
|
||||
min_transformers_version="4.53"),
|
||||
"GlmForCausalLM": _HfExamplesInfo("THUDM/glm-4-9b-chat-hf"),
|
||||
"Glm4ForCausalLM": _HfExamplesInfo("THUDM/GLM-4-9B-0414"),
|
||||
"GlmForCausalLM": _HfExamplesInfo("zai-org/glm-4-9b-chat-hf"),
|
||||
"Glm4ForCausalLM": _HfExamplesInfo("zai-org/GLM-4-9B-0414"),
|
||||
"Glm4MoeForCausalLM": _HfExamplesInfo("zai-org/GLM-4.5",
|
||||
min_transformers_version="4.54"), # noqa: E501
|
||||
"GPT2LMHeadModel": _HfExamplesInfo("openai-community/gpt2",
|
||||
@@ -380,10 +380,10 @@ _MULTIMODAL_EXAMPLE_MODELS = {
|
||||
"FuyuForCausalLM": _HfExamplesInfo("adept/fuyu-8b"),
|
||||
"Gemma3ForConditionalGeneration": _HfExamplesInfo("google/gemma-3-4b-it"),
|
||||
"GraniteSpeechForConditionalGeneration": _HfExamplesInfo("ibm-granite/granite-speech-3.3-2b"), # noqa: E501
|
||||
"GLM4VForCausalLM": _HfExamplesInfo("THUDM/glm-4v-9b",
|
||||
"GLM4VForCausalLM": _HfExamplesInfo("zai-org/glm-4v-9b",
|
||||
trust_remote_code=True,
|
||||
hf_overrides={"architectures": ["GLM4VForCausalLM"]}), # noqa: E501
|
||||
"Glm4vForConditionalGeneration": _HfExamplesInfo("THUDM/GLM-4.1V-9B-Thinking"), # noqa: E501
|
||||
"Glm4vForConditionalGeneration": _HfExamplesInfo("zai-org/GLM-4.1V-9B-Thinking"), # noqa: E501
|
||||
"Glm4v_moeForConditionalGeneration": _HfExamplesInfo("zai-org/GLM-4.5V-Air",
|
||||
is_available_online=False), # noqa: E501
|
||||
"H2OVLChatModel": _HfExamplesInfo("h2oai/h2ovl-mississippi-800m",
|
||||
|
||||
Reference in New Issue
Block a user