Search results

Potential privacy / IP leakage from datastore-backed speculative decoding

Status: Open.

#36 In FasterDecoding/REST;

· fastdecoding opened on May 6, 2026

[llama3 branch] DraftRetriever Reader panics when there is exactly one matching suffix

Status: Open.

#34 In FasterDecoding/REST;

· xbfs opened on Mar 4, 2026

[BUG?] There appears to be a bug in DraftRetriever's core search code.

Status: Open.

#33 In FasterDecoding/REST;

· 2065037421 opened on Jan 2, 2026

thread '<unnamed>' panicked at 'called Option::unwrap() on a None value'

Status: Open.

#31 In FasterDecoding/REST;

· homehappily opened on Apr 28, 2025

question about token generation

Status: Open.

#30 In FasterDecoding/REST;

· YouthLiuYS opened on Apr 27, 2025

Problem with draftretriever.search function behavior

Status: Open.

#29 In FasterDecoding/REST;

· qingyiaaaaa opened on Feb 20, 2025

Lama2 70B-chat is not supported?

Status: Open.

#22 In FasterDecoding/REST;

· Chlience opened on Oct 17, 2024

Tensor parallelism

Status: Open.

#21 In FasterDecoding/REST;

· lethean1 opened on Oct 11, 2024

about retrieve sequence length

Status: Open.

#20 In FasterDecoding/REST;

· Siegfried-qgf opened on Sep 27, 2024

OSError: failed to fill whole buffer

Status: Open.

#19 In FasterDecoding/REST;

· Siegfried-qgf opened on Sep 25, 2024

How to adjust the maximum token lengths when drafting?

Status: Open.

#18 In FasterDecoding/REST;

· zomss opened on Sep 7, 2024

If I want to change REST code to support multi-batch inference, what needs to be changed?

Status: Open.

#16 In FasterDecoding/REST;

· yangbohust opened on May 20, 2024