Skip to content

ASC-Competition/ASC24-LLM-inference-optimization

View on GitHub ↗Related repositories →

The dataset and baseline code for ASC24 LLM inference optimization challenge.

active 2023-12-152024-11-14 (UTC)

Complete coverage26,377 / 26,377 hourly files (100%) · 2 absent upstream2023-08-152026-08-18 (UTC)
Events
83
Pushes
4
Pull requests
0
Issues
16
Stars
35
Forks
10

Activity over time

Daily event counts in the loaded window

Line chart, 336 days from 2023-12-15 to 2024-11-14. Pushes: 4 total, peak 2 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 16 total, peak 7 in a day. Comments: 17 total, peak 8 in a day. Stars: 35 total, peak 4 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Top contributors

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

ContributorContributionsPushesPRsComments
ASC-Competition11407
Thewillman7005
Traveller20016000
Lvjinhong6002
Kevin-shihello-world5003
Harold-QIU2000

Recent activity

Latest issues, pull requests and releases

  • Issue#2Harold-QIU2024-01-14 01:41
    The weights file link provided is not consistent with the baseline code
  • Issue#3Lvjinhong2023-12-31 08:54
    Regarding Specific Guidance on Building an Inference System
  • Issue comment#8Thewillman2023-12-29 06:45
    About total token
  • Issue comment#8Thewillman2023-12-29 06:40
    About total token
  • Issue comment#8Lvjinhong2023-12-27 09:24
    About total token
  • Issue comment#8Thewillman2023-12-25 11:47
    About total token
  • Issue comment#8Thewillman2023-12-25 11:47
    About total token
  • Issue comment#8Thewillman2023-12-25 11:46
    About total token
  • Issue#8Thewillman2023-12-25 11:09
    About total token
  • Issue#7Traveller20012023-12-22 02:49
    About total token
  • Issue comment#7ASC-Competition2023-12-22 01:22
    About total token
  • Issue#7Traveller20012023-12-21 09:31
    About total token
  • Issue#6Kevin-shihello-world2023-12-21 03:04
    Want a discussion area here in this repository for we competitors to discuss and learn from each other.
  • Issue#5Thewillman2023-12-19 12:00
    How about some ml-based method?
  • Issue#3Lvjinhong2023-12-19 11:39
    Regarding Specific Guidance on Building an Inference System
  • Issue#4Traveller20012023-12-19 11:18
    Whether to allow methods that use multiple model copies
  • Issue#4Traveller20012023-12-19 11:18
    Whether to allow methods that use multiple model copies
  • Issue#4Traveller20012023-12-19 11:18
    Whether to allow methods that use multiple model copies
  • Issue#3Lvjinhong2023-12-19 11:10
    Regarding Specific Guidance on Building an Inference System
  • Issue comment#1ASC-Competition2023-12-19 10:41
    Is that we should fully use all 70b parameter? I mean can we use some skill like discrimination to make it smaller Or use another some small model to help its inference Or separate the whole big model into some small part and use Moe to help it move faster?
  • Issue comment#3ASC-Competition2023-12-19 10:21
    Regarding Specific Guidance on Building an Inference System
  • Issue#4Traveller20012023-12-19 03:43
    Whether to allow methods that use multiple model copies
  • Issue comment#1Kevin-shihello-world2023-12-18 11:19
    Is that we should fully use all 70b parameter? I mean can we use some skill like discrimination to make it smaller Or use another some small model to help its inference Or separate the whole big model into some small part and use Moe to help it move faster?
  • Issue comment#3Lvjinhong2023-12-18 11:06
    Regarding Specific Guidance on Building an Inference System
  • Issue comment#1ASC-Competition2023-12-18 09:30
    Is that we should fully use all 70b parameter? I mean can we use some skill like discrimination to make it smaller Or use another some small model to help its inference Or separate the whole big model into some small part and use Moe to help it move faster?

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 35 stars here means stars gained during the window, not the repo's star count.