LoreonLabsPlatform
DocsHome
  • Overview

Intelligence

  • Markets
  • Builders
  • Research
  • Ecosystems
  • Launchpads
  • Search
Ecosystems

Other

SpecForge

Train speculative decoding models effortlessly and port them smoothly to SGLang serving.

OtherEmerging
GitHubWebsite
Stars
—
Forks
—
Contributors
8
Last push
6mo ago

Recent commits

Latest commits.

  • add missing layers_to_output_hidden_states in qwen3 moe (#351)
    a77b9deYubo Wang6mo ago
  • added model-download-dir (#347)
    5c355b8Shenggui Li7mo ago
  • Feature/online train use hf backend optimize GPU usage (#346)
    94de9f8jiapingW7mo ago
  • added profiling (#344)
    3bca52cShenggui Li7mo ago
  • grouped args for better reference (#343)
    44d5c62Shenggui Li7mo ago
updated benchmark docs (#340)
3e0cda0Shenggui Li7mo ago
  • fix mmstart benchmrk (#334)
    5c43694jiapingW7mo ago
  • Remove full hidden states capturing in custom backend (#337)
    e7b3716Yubo Wang7mo ago

    • Top contributors

    Builders behind this project.

    FrankLeeeee
    44 commits
    fzyzcjy
    12 commits
    jiapingW
    11 commits
    yubofredwang
    9 commits
    sleepcoo
    8 commits
    shimizust
    7 commits
    Ximingwang-09
    6 commits
    ZhengHSI
    6 commits