[Bugfix][MRV2] Keep input_ids on later PP stages for raw-token models (#61016)

Signed-off-by: booth-algo <kevinlauofficial01@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Kevin H. Luu <khluu000@gmail.com>
This commit is contained in:
Kevin
2026-10-15 12:54:21 -07:00
committed by GitHub
co-authored by Claude Opus 5.5 Kevin H. Luu
parent d486219fc2
commit d12ea8e309
6 changed files with 51 additions and 5 deletions
@@ -7,7 +7,7 @@ chatml_jinja_path = VLLM_PATH / "examples/template_chatml.jinja"
assert chatml_jinja_path.exists()
def run_and_test_dummy_opt_api_server(model, tp=1):
def run_and_test_dummy_opt_api_server(model, tp=1, extra_args=(), env_dict=None):
# the model is registered through the plugin
server_args = [
"--gpu-memory-utilization",
@@ -20,8 +20,9 @@ def run_and_test_dummy_opt_api_server(model, tp=1):
"dummy",
"-tp",
f"{tp}",
*extra_args,
]
with RemoteOpenAIServer(model, server_args) as server:
with RemoteOpenAIServer(model, server_args, env_dict=env_dict) as server:
client = server.get_client()
completion = client.chat.completions.create(
model=model,