[Perf] Reuse verified H3 prepared weights across model loads - #593
Draft
yangzhuxinyzx wants to merge 6 commits into
Draft
yangzhuxinyzx wants to merge 6 commits into
yangzhuxinyzx wants to merge 6 commits into
Conversation
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>
Signed-off-by: yangzhuxinyzx <153831768+yangzhuxinyzx@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Purpose
Reduce H3 cold-load read/write amplification across original, INT8 and LoRA workflows. Base: fe67339.
Test Plan
Preserve exact weights/layouts, test atomic cache publication, invalidation, leases and disk limits, then measure real first/repeated loads and generated output on the allocated remote four-card V100 host. Do not interrupt active user generations.
Test Result
Baseline and design contract recorded in docs/design/minimax_h3/COLD_LOADING.md. Implementation and acceptance in progress; no new speed claim.