ggml-org/llama.cpp

common: download from URL, improve parallel download progress status

已关闭

#6,537 创建于 2024年4月8日

 (7 条评论) (0 个反应) (1 位负责人)C++ (21,764 个派生)batch import
enhancementgood first issuehelp wantedsplit

仓库指标

星标
 (124,134 个星标)
PR 合并指标
 (平均合并 6天 8小时) (30 天内合并 389 个 PR)

描述

Context

When downloading a sharded model, files are downloaded in parallel, it was added in:

  • #6192

The progressions of each download conflict: image

Need to properly implement CURLOPT_NOPROGRESS for parallel download.

Example in #6515:

main --hf-repo ggml-org/models \
  --hf-file grok-1/grok-1-q4_0-00001-of-00009.gguf \
  --model   models/grok-1-q4_0-00001-of-00009.gguf \
  -ngl 64
   --prompt "I believe the meaning of life is"

贡献者指南