github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #4908] integration with microsoft/aici for deterministic prompt outputs #28861

New Issue
Open
opened 2026-04-22 07:24:59 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-04-22 07:24:59 -05:00
Owner
Copy Link

Originally created by @bvelker on GitHub (Jun 7, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/4908

Is there a roadmap for combining with microsoft's open source AI Control Interface(AICI) project? AICI affords arbitrary deterministic prompt outputs constrained by a regex expression(and/or other techniques). When developers are deciding between proprietary and open source llm solutions, prompt control seems like it's one of the foremost issues(this is why people choose the "smarter" llms). The advantages to the AICI approach to prompting seems like it's ideas will make AICI(or something else like it) a clear long term winning move for the open source space, affording greater output control and enabling very focused problem solving.

https://github.com/microsoft/aici

As an aside, I think the Agents are currently an overhyped workflow. Even current sota llms can't be consistently trusted to generally solve problems recursively. Given this glaring issue, we need to craft workflows that can reliably squeeze every bit of problem solving out of these models. The AICI approach seems to REALLY help with the reliability. It's like function calling, but even more flexible. Just some thoughts...

Originally created by @bvelker on GitHub (Jun 7, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/4908 Is there a roadmap for combining with microsoft's open source AI Control Interface(AICI) project? AICI affords arbitrary deterministic prompt outputs constrained by a regex expression(and/or other techniques). When developers are deciding between proprietary and open source llm solutions, prompt control seems like it's one of the foremost issues(this is why people choose the "smarter" llms). The advantages to the AICI approach to prompting seems like it's ideas will make AICI(or something else like it) a clear long term winning move for the open source space, affording greater output control and enabling very focused problem solving. https://github.com/microsoft/aici As an aside, I think the Agents are currently an overhyped workflow. Even current sota llms can't be consistently trusted to generally solve problems recursively. Given this glaring issue, we need to craft workflows that can reliably squeeze every bit of problem solving out of these models. The AICI approach seems to REALLY help with the reliability. It's like function calling, but even more flexible. Just some thoughts...
GiteaMirror added the feature request label 2026-04-22 07:24:59 -05:00
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#28861
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?