github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-15 12:48:05 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #10374] no caching and tokenizing redundancy #6818

New Issue
Closed
opened 2026-04-12 18:36:56 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-04-12 18:36:56 -05:00
Owner
Copy Link

Originally created by @Abdulrahman392011 on GitHub (Apr 22, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10374

previous messages need to be tokenized and extract the embeddings along with the last message before pushing them into the language model. this needs to happen every single time. there is a framework that does the caching of the previous messages to prevent the redundancy and save resources and speed up the first response token in large context windows. it's called GPT CACHE.

from my understanding it is compatible with ollama and MIT license. can you guys look into implementing it into ollama API.

Originally created by @Abdulrahman392011 on GitHub (Apr 22, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10374 previous messages need to be tokenized and extract the embeddings along with the last message before pushing them into the language model. this needs to happen every single time. there is a framework that does the caching of the previous messages to prevent the redundancy and save resources and speed up the first response token in large context windows. it's called GPT CACHE. from my understanding it is compatible with ollama and MIT license. can you guys look into implementing it into ollama API.
GiteaMirror added the feature request label 2026-04-12 18:36:56 -05:00
GiteaMirror commented 2026-04-12 18:36:58 -05:00
Author
Owner
Copy Link

@Abdulrahman392011 commented on GitHub (Apr 22, 2025):

please correct me if I'm wrong about this. cause it seems unlikely that this hasn't been addressed prior to today. but the way the context window is set is somewhat obscure to me.

please note that I'm using the python API. perhaps this is different for the ollama run command in terminal.

<!-- gh-comment-id:2822292328 --> @Abdulrahman392011 commented on GitHub (Apr 22, 2025): please correct me if I'm wrong about this. cause it seems unlikely that this hasn't been addressed prior to today. but the way the context window is set is somewhat obscure to me. please note that I'm using the python API. perhaps this is different for the ollama run command in terminal.
GiteaMirror commented 2026-04-12 18:37:00 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Apr 22, 2025):

If you mean GPT cache from zilliz, it's a service that sits between the client and the LLM. There's no need to implement it in ollama, a user just needs to set up the cache server and configure server/client.

ALso, ollama does internal caching for prompts.

<!-- gh-comment-id:2822299352 --> @rick-github commented on GitHub (Apr 22, 2025): If you mean GPT cache from zilliz, it's a service that sits between the client and the LLM. There's no need to implement it in ollama, a user just needs to set up the cache server and configure server/client. ALso, ollama does internal caching for prompts.
GiteaMirror commented 2026-04-12 18:37:02 -05:00
Author
Owner
Copy Link

@Abdulrahman392011 commented on GitHub (Apr 22, 2025):

just to confirm, I am using ollama python API. it automatically manage the previous messages and there is no redundancy in processing the previous messages with every new message.

<!-- gh-comment-id:2822306523 --> @Abdulrahman392011 commented on GitHub (Apr 22, 2025): just to confirm, I am using ollama python API. it automatically manage the previous messages and there is no redundancy in processing the previous messages with every new message.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#6818
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?