From e33a50480ddc9db9789d5f15ae10f92ebc6be4c6 Mon Sep 17 00:00:00 2001
From: Matthew Farrellee <matt@cs.wisc.edu>
Date: Thu, 24 Jul 2025 11:28:37 -0400
Subject: [PATCH] fix: starter template and litellm backward compat conflict
 for openai (#2885)

# What does this PR do?

openai/models.py has backward compat entries for litellm model names.
the starter template includes these in the list of registered models.
the inclusion results in duplicate model registrations.

the backward compat is no longer necessary.

## Test Plan

ci
---
 .../remote/inference/openai/models.py         |  7 -----
 llama_stack/templates/ci-tests/run.yaml       | 29 -------------------
 llama_stack/templates/starter/run.yaml        | 29 -------------------
 3 files changed, 65 deletions(-)

diff --git a/llama_stack/providers/remote/inference/openai/models.py b/llama_stack/providers/remote/inference/openai/models.py
index 37bee57de..28d0c4b41 100644
--- a/llama_stack/providers/remote/inference/openai/models.py
+++ b/llama_stack/providers/remote/inference/openai/models.py
@@ -12,11 +12,6 @@ from llama_stack.providers.utils.inference.model_registry import (
 )
 
 LLM_MODEL_IDS = [
-    # the models w/ "openai/" prefix are the litellm specific model names.
-    # they should be deprecated in favor of the canonical openai model names.
-    "openai/gpt-4o",
-    "openai/gpt-4o-mini",
-    "openai/chatgpt-4o-latest",
     "gpt-3.5-turbo-0125",
     "gpt-3.5-turbo",
     "gpt-3.5-turbo-instruct",
@@ -43,8 +38,6 @@ class EmbeddingModelInfo:
 
 
 EMBEDDING_MODEL_IDS: dict[str, EmbeddingModelInfo] = {
-    "openai/text-embedding-3-small": EmbeddingModelInfo(1536, 8192),
-    "openai/text-embedding-3-large": EmbeddingModelInfo(3072, 8192),
     "text-embedding-3-small": EmbeddingModelInfo(1536, 8192),
     "text-embedding-3-large": EmbeddingModelInfo(3072, 8192),
 }
diff --git a/llama_stack/templates/ci-tests/run.yaml b/llama_stack/templates/ci-tests/run.yaml
index 3555c2d5e..1396d54a8 100644
--- a/llama_stack/templates/ci-tests/run.yaml
+++ b/llama_stack/templates/ci-tests/run.yaml
@@ -785,21 +785,6 @@ models:
   provider_id: ${env.ENABLE_RUNPOD:=__disabled__}
   provider_model_id: Llama3.2-3B
   model_type: llm
-- metadata: {}
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/gpt-4o
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/gpt-4o
-  model_type: llm
-- metadata: {}
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/gpt-4o-mini
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/gpt-4o-mini
-  model_type: llm
-- metadata: {}
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/chatgpt-4o-latest
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/chatgpt-4o-latest
-  model_type: llm
 - metadata: {}
   model_id: ${env.ENABLE_OPENAI:=__disabled__}/gpt-3.5-turbo-0125
   provider_id: ${env.ENABLE_OPENAI:=__disabled__}
@@ -870,20 +855,6 @@ models:
   provider_id: ${env.ENABLE_OPENAI:=__disabled__}
   provider_model_id: o4-mini
   model_type: llm
-- metadata:
-    embedding_dimension: 1536
-    context_length: 8192
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/text-embedding-3-small
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/text-embedding-3-small
-  model_type: embedding
-- metadata:
-    embedding_dimension: 3072
-    context_length: 8192
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/text-embedding-3-large
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/text-embedding-3-large
-  model_type: embedding
 - metadata:
     embedding_dimension: 1536
     context_length: 8192
diff --git a/llama_stack/templates/starter/run.yaml b/llama_stack/templates/starter/run.yaml
index 32ee0ea21..c38933f98 100644
--- a/llama_stack/templates/starter/run.yaml
+++ b/llama_stack/templates/starter/run.yaml
@@ -785,21 +785,6 @@ models:
   provider_id: ${env.ENABLE_RUNPOD:=__disabled__}
   provider_model_id: Llama3.2-3B
   model_type: llm
-- metadata: {}
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/gpt-4o
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/gpt-4o
-  model_type: llm
-- metadata: {}
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/gpt-4o-mini
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/gpt-4o-mini
-  model_type: llm
-- metadata: {}
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/chatgpt-4o-latest
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/chatgpt-4o-latest
-  model_type: llm
 - metadata: {}
   model_id: ${env.ENABLE_OPENAI:=__disabled__}/gpt-3.5-turbo-0125
   provider_id: ${env.ENABLE_OPENAI:=__disabled__}
@@ -870,20 +855,6 @@ models:
   provider_id: ${env.ENABLE_OPENAI:=__disabled__}
   provider_model_id: o4-mini
   model_type: llm
-- metadata:
-    embedding_dimension: 1536
-    context_length: 8192
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/text-embedding-3-small
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/text-embedding-3-small
-  model_type: embedding
-- metadata:
-    embedding_dimension: 3072
-    context_length: 8192
-  model_id: ${env.ENABLE_OPENAI:=__disabled__}/openai/text-embedding-3-large
-  provider_id: ${env.ENABLE_OPENAI:=__disabled__}
-  provider_model_id: openai/text-embedding-3-large
-  model_type: embedding
 - metadata:
     embedding_dimension: 1536
     context_length: 8192