chore: update model catalog from bot issues

[github-actions]

Automated daily batch of model catalog updates from bot issues.

Included issues

• Closes [BOT ISSUE] Bedrock: add missing Moonshot Kimi K3 models (moonshotai.kimi-k3 + geo variants) #1271: [BOT ISSUE] Bedrock: add missing Moonshot Kimi K3 models (moonshotai.kimi-k3 + geo variants)
• Closes [BOT ISSUE] Google: add missing gemini-3.8-live and gemini-3.8-live-extended-thinking #1279: [BOT ISSUE] Google: add missing gemini-3.8-live and gemini-3.8-live-extended-thinking

Summary

Issue	Provider	Primary model	Changed models	Added models	Updated models	Verification sources
#1271	bedrock	moonshotai.kimi-k3	moonshotai.kimi-k3 us.moonshotai.kimi-k3 global.moonshotai.kimi-k3	moonshotai.kimi-k3 us.moonshotai.kimi-k3 global.moonshotai.kimi-k3	None	1 2
#1279	google	gemini-3.8-live	gemini-3.8-live gemini-3.8-live-extended-thinking	gemini-3.8-live gemini-3.8-live-extended-thinking	None	1 2 3 4 5

Verified metadata

#1271: [BOT ISSUE] Bedrock: add missing Moonshot Kimi K3 models (moonshotai.kimi-k3 + geo variants)

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
moonshotai.kimi-k3	Kimi K3		bedrock	converse	chat	input=1000000, output=not provided	in/out=3/15 per 1M; cache read=0.3 per 1M; cache write=3.75 per 1M	multimodal=true
us.moonshotai.kimi-k3	Kimi K3	moonshotai.kimi-k3	bedrock	converse	chat	input=1000000, output=not provided	in/out=3/15 per 1M; cache read=0.3 per 1M; cache write=3.75 per 1M	parent=moonshotai.kimi-k3; multimodal=true
global.moonshotai.kimi-k3	Kimi K3	moonshotai.kimi-k3	bedrock	converse	chat	input=1000000, output=not provided	in/out=3/15 per 1M; cache read=0.3 per 1M; cache write=3.75 per 1M	parent=moonshotai.kimi-k3; multimodal=true

Verification notes

No LLM verification step ran; model metadata was already complete in the issue.

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
moonshotai.kimi-k3	max_output_tokens	n/a	131072	moonshotai.kimi-k3
us.moonshotai.kimi-k3	max_output_tokens	n/a	131072	us.moonshotai.kimi-k3
us.moonshotai.kimi-k3	input_cost_per_mil_tokens	3	3.3	us.moonshotai.kimi-k3
us.moonshotai.kimi-k3	output_cost_per_mil_tokens	15	16.5	us.moonshotai.kimi-k3
us.moonshotai.kimi-k3	input_cache_read_cost_per_mil_tokens	0.3	0.33	us.moonshotai.kimi-k3
us.moonshotai.kimi-k3	input_cache_write_cost_per_mil_tokens	3.75	4.125	us.moonshotai.kimi-k3
global.moonshotai.kimi-k3	max_output_tokens	n/a	131072	global.moonshotai.kimi-k3

#1279: [BOT ISSUE] Google: add missing gemini-3.8-live and gemini-3.8-live-extended-thinking

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
gemini-3.8-live	Gemini 3.8 Live		google	google	live	input=131072, output=65536	in/out=0.75/4.5 per 1M	multimodal=true
gemini-3.8-live-extended-thinking	Gemini 3.8 Live Extended Thinking		google	google	live	input=131072, output=65536	in/out=0.75/4.5 per 1M	multimodal=true; reasoning=true

Verification notes

Verification Checklist

sync_models vs proposed update

sync_models cross-check found no pricing/token discrepancies for the changed models.

No sync_models discrepancies were noted.

[github-actions]

Codex (@codex) review

[vercel]

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

Project	Deployment	Actions	Updated
ai-proxy	Ready	Preview	Sep 22, 2026 12:48pm UTC

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 26f014ba86

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]

Move endpoint-routing additions to the gateway repository

These AvailableEndpointTypes entries modify provider-routing metadata in the deprecated proxy, even though the repository guidance directs all model-routing work to the parent gateway/ repository and permits only model_list.json catalog changes here. The entries are also unnecessary locally because each new model’s format already resolves to the same default endpoint type; keep the catalog additions but remove these index.ts changes and apply any required routing update in ../gateway/.

AGENTS.md reference: AGENTS.md:L10-L15

Useful? React with 👍 / 👎.