Skip to content

incorporating-admission-controler - #271

Merged
masalim2 merged 31 commits into
firstv2from
firstv2-admission-controler
Aug 20, 2026
Merged

masalim2 merged 31 commits into
firstv2from
firstv2-admission-controler

Conversation

@bcote-anl

Copy link
Copy Markdown
Collaborator

Incorporating AdmissionControler to OpenAI routes

This PR includes:

  • new FIRST error classes
  • AdmissionControler dependency
  • service/orchestration.py file to extract target backend given model deployments
  • submit_inference.py placeholder
  • routing to specific deployment as well as federated routing

@bcote-anl
bcote-anl requested a review from masalim2 July 27, 2026 20:46
@bcote-anl bcote-anl changed the title adding openai chat/completions pydantic class incorporating-admission-controler Jul 27, 2026

@masalim2 masalim2 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

This is fantastic, thank you for building this out! Many of these comments are more forward-looking considerations for future PRs; nothing is blocking the merge of this step! Let me know what you think!

Comment thread packages/common/first_common/errors.py Outdated
Comment thread packages/gateway/first_gateway/apiserver/routes/openai/deployments.py Outdated
Comment thread packages/gateway/first_gateway/apiserver/routes/openai/deployments.py Outdated
Comment thread packages/gateway/first_gateway/apiserver/routes/openai/federated.py Outdated
Comment thread packages/gateway/first_gateway/apiserver/routes/openai/deployments.py Outdated
Comment thread packages/gateway/first_gateway/services/orchestration.py Outdated
Comment thread packages/gateway/first_gateway/services/orchestration.py Outdated
Comment thread packages/gateway/first_gateway/services/orchestration.py
Comment thread packages/gateway/first_gateway/services/orchestration.py Outdated
Comment thread packages/gateway/first_gateway/services/orchestration.py
Comment thread packages/common/first_common/errors.py
Comment thread packages/gateway/first_gateway/services/submit_inference.py
Comment thread packages/gateway/first_gateway/services/submit_inference.py Outdated
Comment thread packages/gateway/first_gateway/services/submit_inference.py
@masalim2

Copy link
Copy Markdown
Collaborator

Looking great! Just a handful of minor comments to address on this PR and I will populate some issues to round up the remaining ideas that we discussed for future PRs!

BenoitCote and others added 19 commits August 20, 2026 10:17
light edits to pydantic models

adding OpenAI Responses pydantic

adding placeholder for litellm router

added first pass on adding openai routes

adding service files to submit inference tasks through litellm

added stramlined pydantic models for OpenAI

removed pydantic classes which have moved to the endpoints/ folder

added not-implemented error

added common base payload and enum for openai endpoints

removed llm_router.py

added first pass on openai routes

added openai router to the main router

added embeddings

added embeddings dependency

added skeleton for embeddings route

empty deployments file with router

removed old unused dev file

removed inference service file since it is outdated

use production policy

switched to RouterConfigDep, fix argument typo

added response_model=None to accomodate StreamingResponse type

added return type to model_dependency

List to list for typing

streamlined pydantic models

added str to the Enum class

fixed typo in the Enum class

added alias check

streamlined dependency to resolve model

added service orchestration file for reusable functions to find the backend

added errors

added admission controler dependency

added admission controler to API route arguments

fixed id typo. made fcts sync

added placeholder for submit inference call function

added numpy

added numpy

point to submit_inference

added deployment routes
@bcote-anl
bcote-anl force-pushed the firstv2-admission-controler branch from 9b60ad6 to 1798d8b Compare August 20, 2026 17:17
@bcote-anl
bcote-anl marked this pull request as ready for review August 20, 2026 20:41
@masalim2
masalim2 merged commit acbfd86 into firstv2 Aug 20, 2026
2 checks passed
@bcote-anl
bcote-anl deleted the firstv2-admission-controler branch August 25, 2026 16:44
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.

2 participants