Skip to content

[Model][Last/N] Improve all pooling task | Generate runner supports using embed and token_embed tasks. #30672

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
noooop wants to merge 5 commits into
base: main
Choose a base branch
from
+44 −32
Draft

[Model][Last/N] Improve all pooling task | Generate runner supports using embed and token_embed tasks. #30672

Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
0346166
init
noooop Dec 15, 2025
7873b25
Merge branch 'main' into Prompt_Hidden_States
noooop Dec 15, 2025
64da7a6
very dirty fix
noooop Dec 15, 2025
4b482ac
fix
noooop Dec 15, 2025
e3b1dd9
fix
noooop Dec 15, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 11 additions & 2 deletions tests/models/language/pooling/test_extract_hidden_states.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
import pytest
import torch

from vllm import TokensPrompt
from vllm import SamplingParams, TokensPrompt


@pytest.mark.parametrize(
Expand All @@ -19,7 +19,7 @@ def test_extract_hidden_states(hf_runner, vllm_runner, model: str):
model,
max_model_len=128,
enforce_eager=True,
runner="pooling",
runner="generate",
enable_prefix_caching=True,
) as vllm_model:
pooling_outputs = vllm_model.llm.encode(
Expand Down Expand Up @@ -55,3 +55,12 @@ def test_extract_hidden_states(hf_runner, vllm_runner, model: str):
for n, output in zip(n_prompt_tokens, pooling_outputs):
assert len(output.prompt_token_ids) == n
assert output.num_cached_tokens > 0

# Support generate text and returning Prompt Hidden States
generate_outputs = vllm_model.generate(
prompts=[TokensPrompt(prompt_token_ids=t) for t in token_prompts],
sampling_params=SamplingParams(max_tokens=1),
)
for n, output in zip(n_prompt_tokens, generate_outputs):
assert len(output.prompt_token_ids) == n
assert output.num_cached_tokens > 0