[GLM-OCR] GLM-OCR with MTP Support (#33005)

Signed-off-by: zRzRzRzRzRzRzR <2448370773@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Yuxuan Zhang
2026-01-26 22:24:43 +08:00
committed by GitHub
parent dcd80206b7
commit bb17e8f11c
14 changed files with 873 additions and 8 deletions

View File

@@ -91,6 +91,19 @@ MODEL_CONFIGS: dict[str, dict[str, Any]] = {
"use_processor": True,
"question": "What is the content of each image?",
},
"glm_ocr": {
"model_name": "zai-org/GLM-OCR",
"interface": "llm_generate",
"max_model_len": 131072,
"max_num_seqs": 2,
"sampling_params": {
"temperature": 0.0,
"max_tokens": 256,
"stop_token_ids": None,
},
"use_processor": True,
"question": "Text Recognition:",
},
"keye_vl": {
"model_name": "Kwai-Keye/Keye-VL-8B-Preview",
"interface": "llm_generate",