auto-round
[risky]reduce ram usage of loading moe without monkey patch and fix qwen3-flash-next group_size issue
#2291
Merged

Commits
  • update
    wenhuach21 committed 52 days ago
  • [pre-commit.ci] auto fixes from pre-commit.com hooks
    pre-commit-ci[bot] committed 52 days ago
  • update
    wenhuach21 committed 52 days ago
  • update
    wenhuach21 committed 52 days ago
  • Merge branch 'main' into fix_moe_loading
    wenhuach21 committed 52 days ago
  • [pre-commit.ci] auto fixes from pre-commit.com hooks
    pre-commit-ci[bot] committed 52 days ago
  • disable all moe replacements
    wenhuach21 committed 52 days ago
  • Merge branch 'fix_moe_loading' of https://github.com/intel/auto-round into fix_moe_loading
    wenhuach21 committed 52 days ago
  • fix moe gguf
    wenhuach21 committed 51 days ago
  • fix: restore dedicated MoE replacements
    Copilot committed 51 days ago
  • Merge branch 'main' into fix_moe_loading
    wenhuach21 committed 51 days ago
  • try to fix ut
    wenhuach21 committed 51 days ago
  • Merge branch 'fix_moe_loading' of https://github.com/intel/auto-round into fix_moe_loading
    wenhuach21 committed 51 days ago
  • Merge branch 'main' into fix_moe_loading
    wenhuach21 committed 51 days ago
  • try to fix glm5.3
    wenhuach21 committed 51 days ago
  • update
    wenhuach21 committed 51 days ago
  • fix qwen3.8-flash-next issue
    wenhuach21 committed 51 days ago
  • [pre-commit.ci] auto fixes from pre-commit.com hooks
    pre-commit-ci[bot] committed 51 days ago
  • try to fix ut
    wenhuach21 committed 51 days ago
  • Merge branch 'fix_moe_loading' of https://github.com/intel/auto-round into fix_moe_loading
    wenhuach21 committed 51 days ago
  • [pre-commit.ci] auto fixes from pre-commit.com hooks
    pre-commit-ci[bot] committed 51 days ago
  • try to fix ut
    wenhuach21 committed 51 days ago
  • Merge branch 'fix_moe_loading' of https://github.com/intel/auto-round into fix_moe_loading
    wenhuach21 committed 51 days ago
  • [pre-commit.ci] auto fixes from pre-commit.com hooks
    pre-commit-ci[bot] committed 51 days ago
  • refine
    wenhuach21 committed 51 days ago
  • Merge branch 'fix_moe_loading' of https://github.com/intel/auto-round into fix_moe_loading
    wenhuach21 committed 51 days ago
  • fix ut
    wenhuach21 committed 51 days ago
Loading