Skip to content

Conversation

@wuhuxiao
Copy link
Contributor

@wuhuxiao wuhuxiao commented Dec 3, 2025

Purpose

Context Independent Cache is all you need.
Use chunk cache hit to speed up TTFT.
What this PR does / why we need it?

Modifications

Does this PR introduce any user-facing change?

Test

ENABLE_SPARSE=True
DATA_DIR=/home/data/kv_cache
MODEL_PATH=/home/models/mistralai/Mistral-7B-Instruct-v0.2
BLEND_DATASET_PATH=/home/datasets/LongBench/data/2wikimqa.jsonl
python offline_inference_blend.py
image

How was this patch tested?

Copy link
Contributor

@mag1c-h mag1c-h left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please submit your pull request to the develop branch, the v1 branch is no longer accepting any pull requests.

@ygwpz ygwpz deleted the branch ModelEngine-Group:dev-ucm-v1 December 4, 2025 03:08
@ygwpz ygwpz closed this Dec 4, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants