[Bugfix] Add bot_task option of think_recaption for hunyuanimage3 it2i (#3551)

Signed-off-by: zengchuang <zengchuang3@huawei.com>

examples/offline_inference/hunyuan_image3/end2end.py

@@ -136,7 +136,7 @@ def main():
     if task not in _TASK_PRESETS:
         valid_bot_tasks = {
             "text2img": ["think", "recaption", "vanilla"],
-            "img2img": ["think", "recaption"],
+            "img2img": ["think", "recaption", "think_recaption"],
             "img2text": ["auto"],
             "text2text": ["auto"],
         }[args.modality]

tests/diffusion/models/hunyuan_image3/test_prompt_utils.py

@@ -78,6 +78,7 @@ def test_available_tasks_covers_all_modalities():
         "i2t",
         "it2i_think",
         "it2i_recaption",
+        "it2i_think_recaption",
         "t2i_think",
         "t2i_recaption",
         "t2i_vanilla",
@@ -99,6 +100,7 @@ def test_resolve_stop_token_ids_uses_answer_for_generation_tasks():
         "i2t",
         "it2i_think",
         "it2i_recaption",
+        "it2i_think_recaption",
         "t2i_think",
         "t2i_recaption",
     ],
@@ -125,7 +127,7 @@ def test_build_prompt_string_structure_chat_template(task: str):
     # documentation, so substring index() catches the wrong occurrence -- use
     # endswith() which directly captures "trigger is at the tail" (the Part A
     # fix: trigger goes AFTER `Assistant: `, not before user_prompt).
-    if task in ("it2i_think", "t2i_think"):
+    if task in ("it2i_think", "t2i_think", "it2i_think_recaption"):
         assert s.endswith("Assistant: <think>"), (
             f"Trigger <think> must be appended right after `Assistant: ` (Part A fix). Got tail: ...{s[-40:]!r}"
         )

vllm_omni/diffusion/models/hunyuan_image3/prompt_utils.py

@@ -51,6 +51,7 @@ _TASK_PRESETS: dict[str, tuple[str, str | None, str | None]] = {
     "i2t": ("en_unified", None, None),
     "it2i_think": ("en_unified", "think", "<think>"),
     "it2i_recaption": ("en_unified", "recaption", "<recaption>"),
+    "it2i_think_recaption": ("en_unified", "think_recaption", "<think>"),
     "t2i": ("en_unified", "image", None),
     "t2i_vanilla": ("en_vanilla", "image", None),
     "t2i_think": ("en_unified", "think", "<think>"),