Skip to content

scalable and robust tree-based speculative decoding algorithm

active 2024-02-292026-03-08 (UTC)

Partial coverage20,536 / 26,292 hourly files (78%) · 2 absent upstream · 5,753 failed, retryable2023-08-152026-08-14 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
331
Pushes
68
Pull requests
0
Issues
8
Stars
203
Forks
30

Activity over time

Daily event counts in the loaded window

Line chart, 739 days from 2024-02-29 to 2026-03-08. Pushes: 68 total, peak 13 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 8 total, peak 2 in a day. Comments: 18 total, peak 5 in a day. Stars: 203 total, peak 39 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
dreaming-panda786809
Tomorrowdawn5003
cyLi-Tiger3001
poedator2000
singularity-s02002
lethean11001
tchaton1000
briskerkazoos1001
preminstrel1001

Recent activity

Latest issues, pull requests and releases

  • Issue comment#16dreaming-panda2024-10-27 03:34
    Work On CPU
  • Issue comment#11lethean12024-10-13 03:25
    The support on vLLM?
  • Issue comment#13briskerkazoos2024-05-15 04:32
    Reproducibility: the tree_search generates too small tree
  • Issue comment#13singularity-s02024-04-28 16:29
    Reproducibility: the tree_search generates too small tree
  • Issue comment#13singularity-s02024-04-28 16:24
    Reproducibility: the tree_search generates too small tree
  • Issue#15cyLi-Tiger2024-04-28 03:16
    Question on tree search algorithm
  • Issue comment#15dreaming-panda2024-04-28 03:07
    Question on tree search algorithm
  • Issue comment#15cyLi-Tiger2024-04-28 02:03
    Question on tree search algorithm
  • Issue comment#15dreaming-panda2024-04-27 16:24
    Question on tree search algorithm
  • Issue#15cyLi-Tiger2024-04-27 09:11
    Question on tree search algorithm
  • Issue comment#13dreaming-panda2024-04-27 05:01
    Reproducibility: the tree_search generates too small tree
  • Issue comment#11dreaming-panda2024-04-21 01:56
    The support on vLLM?
  • Issue comment#10dreaming-panda2024-04-12 06:31
    Is there any benchmark that compares Sequoia against vanilla speculative decoding?
  • Issue comment#10preminstrel2024-04-12 05:37
    Is there any benchmark that compares Sequoia against vanilla speculative decoding?
  • Issue#4poedator2024-03-17 19:03
    data loading timing and disk use
  • Issue#3tchaton2024-03-14 07:17
    Integration with Lit-GPT
  • Issue comment#2Tomorrowdawn2024-03-10 14:03
    Tensor shape mismatch when computing apply_rotary_pos_emb
  • Issue#2Tomorrowdawn2024-03-10 14:03
    Tensor shape mismatch when computing apply_rotary_pos_emb
  • Issue#1dreaming-panda2024-03-10 13:27
    Error `p.attn_bias_ptr is not correctly aligned` when testing
  • Issue comment#2dreaming-panda2024-03-10 13:22
    Tensor shape mismatch when computing apply_rotary_pos_emb
  • Issue comment#2Tomorrowdawn2024-03-10 13:17
    Tensor shape mismatch when computing apply_rotary_pos_emb
  • Issue comment#2dreaming-panda2024-03-10 13:06
    Tensor shape mismatch when computing apply_rotary_pos_emb
  • Issue comment#2Tomorrowdawn2024-03-10 06:13
    Assert attention_mask.shape[0] == input_length fails
  • Issue#2Tomorrowdawn2024-03-09 08:23
    Assert attention_mask.shape[0] == input_length fails
  • Issue comment#1dreaming-panda2024-03-03 22:04
    Error `p.attn_bias_ptr is not correctly aligned` when testing

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 203 stars here means stars gained during the window, not the repo's star count.