chore: update model catalog from bot issues

[github-actions]

Automated daily batch of model catalog updates from bot issues.

Included issues

• Closes [BOT ISSUE] OpenAI: remove incorrect cached input pricing from gpt-5.5-pro and gpt-5.4-pro #837: [BOT ISSUE] OpenAI: remove incorrect cached input pricing from gpt-5.5-pro and gpt-5.4-pro
• Closes [BOT ISSUE] OpenAI: add deprecation_date for gpt-4o-2024-05-13, o1-2024-12-17, o4-mini-2025-04-16 #838: [BOT ISSUE] OpenAI: add deprecation_date for gpt-4o-2024-05-13, o1-2024-12-17, o4-mini-2025-04-16
• Closes [BOT ISSUE] OpenAI: add deprecation_date for gpt-5-2025-08-07, o3-2025-04-16, gpt-5-mini-2025-08-07 #839: [BOT ISSUE] OpenAI: add deprecation_date for gpt-5-2025-08-07, o3-2025-04-16, gpt-5-mini-2025-08-07

Summary

Issue	Provider	Primary model	Changed models	Added models	Updated models	Verification sources
#837	openai	gpt-5.5-pro	gpt-5.5-pro gpt-5.5-pro-2026-04-23 gpt-5.4-pro gpt-5.4-pro-2026-03-05	None	gpt-5.5-pro gpt-5.5-pro-2026-04-23 gpt-5.4-pro gpt-5.4-pro-2026-03-05	1
#838	openai	gpt-4o-2024-05-13	gpt-4o-2024-05-13 o1-2024-12-17 o4-mini-2025-04-16	None	gpt-4o-2024-05-13 o1-2024-12-17 o4-mini-2025-04-16	1
#839	openai	gpt-5-2025-08-07	gpt-5-2025-08-07 o3-2025-04-16 gpt-5-mini-2025-08-07	None	gpt-5-2025-08-07 o3-2025-04-16 gpt-5-mini-2025-08-07	1

Verified metadata

#837: [BOT ISSUE] OpenAI: remove incorrect cached input pricing from gpt-5.5-pro and gpt-5.4-pro

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
gpt-5.5-pro	GPT-5.5 Pro		openai, azure	openai	chat	input=1050000, output=128000	in/out=30/180 per 1M; cache read=null per 1M	multimodal=true; reasoning=true
gpt-5.5-pro-2026-04-23	GPT-5.5 Pro (2026-04-23)	gpt-5.5-pro	openai, azure	openai	chat	input=1050000, output=128000	in/out=30/180 per 1M; cache read=null per 1M	parent=gpt-5.5-pro; multimodal=true; reasoning=true
gpt-5.4-pro	GPT-5.4 Pro		openai, azure	openai	chat	input=1050000, output=128000	in/out=30/180 per 1M; cache read=null per 1M	multimodal=true; reasoning=true
gpt-5.4-pro-2026-03-05	GPT-5.4 Pro (2026-03-05)	gpt-5.4-pro	openai, azure	openai	chat	input=1050000, output=128000	in/out=30/180 per 1M; cache read=null per 1M	parent=gpt-5.4-pro; multimodal=true; reasoning=true

Verification notes

Verification

sync_models vs proposed update

sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.

Model	Field	Proposed update	sync_models	sync_models source models
gpt-5.5-pro	input_cache_read_cost_per_mil_tokens	n/a	3	gpt-5.5-pro
gpt-5.5-pro-2026-04-23	input_cache_read_cost_per_mil_tokens	n/a	3	gpt-5.5-pro-2026-04-23
gpt-5.4-pro	input_cache_read_cost_per_mil_tokens	n/a	3	gpt-5.4-pro
gpt-5.4-pro-2026-03-05	input_cache_read_cost_per_mil_tokens	n/a	3	gpt-5.4-pro-2026-03-05

#838: [BOT ISSUE] OpenAI: add deprecation_date for gpt-4o-2024-05-13, o1-2024-12-17, o4-mini-2025-04-16

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
gpt-4o-2024-05-13	GPT-4o (2024-05-13)	gpt-4o	openai, azure	openai	chat	input=128000, output=4096	in/out=5/15 per 1M	parent=gpt-4o; date=2026-10-23; multimodal=true
o1-2024-12-17	o1 (2024-12-17)	o1	openai, azure	openai	chat	input=200000, output=100000	in/out=15/60 per 1M; cache read=7.5 per 1M	parent=o1; date=2026-10-23; multimodal=true; reasoning=true
o4-mini-2025-04-16	o4-mini (2025-04-16)	o4-mini	openai, azure	openai	chat	input=200000, output=100000	in/out=1.1/4.4 per 1M; cache read=0.275 per 1M	parent=o4-mini; date=2026-10-23; multimodal=true; reasoning=true

Verification notes

Verification

sync_models vs proposed update

sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.

Model	Field	Proposed update	sync_models	sync_models source models
gpt-4o-2024-05-13	deprecation_date	2026-10-23	n/a	gpt-4o-2024-05-13
o1-2024-12-17	deprecation_date	2026-10-23	n/a	o1-2024-12-17
o4-mini-2025-04-16	deprecation_date	2026-10-23	n/a	o4-mini-2025-04-16

#839: [BOT ISSUE] OpenAI: add deprecation_date for gpt-5-2025-08-07, o3-2025-04-16, gpt-5-mini-2025-08-07

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
gpt-5-2025-08-07	GPT-5 (2025-08-07)	gpt-5	openai, azure	openai	chat	input=272000, output=128000	in/out=1.25/10 per 1M; cache read=0.125 per 1M	parent=gpt-5; date=2026-12-11; multimodal=true; reasoning=true
o3-2025-04-16	o3 (2025-04-16)	o3	openai, azure	openai	chat	input=200000, output=100000	in/out=2/8 per 1M; cache read=0.5 per 1M	parent=o3; date=2026-12-11; multimodal=true; reasoning=true
gpt-5-mini-2025-08-07	GPT-5 mini (2025-08-07)	gpt-5-mini	openai, azure	openai	chat	input=272000, output=128000	in/out=0.25/2 per 1M; cache read=0.025 per 1M	parent=gpt-5-mini; date=2026-12-11; multimodal=true; reasoning=true

Verification notes

Verification

sync_models vs proposed update

sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.

Model	Field	Proposed update	sync_models	sync_models source models
gpt-5-2025-08-07	deprecation_date	2026-12-11	n/a	gpt-5-2025-08-07
o3-2025-04-16	deprecation_date	2026-12-11	n/a	o3-2025-04-16
gpt-5-mini-2025-08-07	deprecation_date	2026-12-11	n/a	gpt-5-mini-2025-08-07

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
ai-proxy	Ready	Preview, Comment	Jun 21, 2026 11:22am

[github-actions]

Codex (@codex) review

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 204f1a644d

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "Codex (@codex) review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "Codex (@codex) address that feedback".

[chatgpt-codex-connector]

Don't advertise Bedrock for OpenAI-format specs

When a request uses one of these newly mapped Bedrock-only models, getModelEndpointTypes now selects a Bedrock secret, but the catalog entries for openai.gpt-5.5, openai.gpt-5.4, and xai.grok-4.3 still declare format: "openai". That path enters fetchOpenAI, which explicitly rejects secret.type === "bedrock", so these models will always fail through the only provider advertised here unless they are routed through a supported Bedrock/Converse format or not exposed as Bedrock yet.

Useful? React with 👍 / 👎.

[github-actions]

Codex Review Response

Issue: Bedrock-only models with format: "openai" will always fail

Status: Fixed

Codex correctly identified that openai.gpt-5.5, openai.gpt-5.4, and xai.grok-4.3 declared format: "openai" while being Bedrock-only models. The proxy's fetchOpenAI function explicitly rejects Bedrock secrets (proxy.ts:2130), so these models would always error at runtime.

Change: Set format to "converse" for all three models in packages/proxy/schema/model_list.json, matching the convention used by every other Bedrock-only third-party model in the catalog (e.g., openai.gpt-oss-*, mistral.*, moonshot.*, zai.*).

No changes were needed in packages/proxy/schema/index.ts.