[Feature][DSpark]: Logprobs adaptive verification (#52242)

Signed-off-by: Naveenraj Kamalakannan <therealnaveenkamal@gmail.com>
Co-authored-by: Benjamin Chislett <bchislett@nvidia.com>
This commit is contained in:
Naveenraj Kamalakannan
2026-08-25 10:22:33 -04:00
committed by GitHub
co-authored by Benjamin Chislett
parent ca29cd48ba
commit d5cadcee86
10 changed files with 61 additions and 25 deletions
-14
View File
@@ -974,20 +974,6 @@ class SamplingParams(
if speculative_config is None:
return
# Adaptive verification compacts logits after the forward pass, while
# compute_topk_scores uses the scheduled layout in cu_num_logits_np.
# TODO(lucas): lift this restriction. The true boundaries exist on device as
# cu_num_logits; cu_num_generated_tokens is only read host-side after
# the logprobs D2H, so it could ride along on that copy.
if (
speculative_config.enable_adaptive_verification
and self.num_logprobs is not None
):
raise ValueError(
"Output logprobs are not supported with DSpark confidence-based "
"verification."
)
# Some sampling parameters are not yet compatible with spec decoding.
if self.min_p > _SAMPLING_EPS or self.logit_bias:
raise VLLMValidationError(