fix: update xAI model metadata for grok-4.20-0309-reasoning

[github-actions]

fix: update xAI model metadata for grok-4.20-0309-reasoning

Closes #545

Source issue: #545

Summary

Field	Value
Provider	xai
Primary model	grok-4.20-0309-reasoning
Changed models	grok-4.20-0309-reasoning
Added models	None
Updated models	grok-4.20-0309-reasoning
Verification sources	1 2 3 4

Verified metadata

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
grok-4.20-0309-reasoning	Grok 4.20 Reasoning (0309)		xAI	openai	chat	input=1000000, output=2000000	in/out=1.25/2.5 per 1M; cache read=0.2 per 1M	multimodal=true; reasoning=true

Verification notes

Verification

Official sources consulted

• https://docs.x.ai/developers/models — Verified: input pricing ($1.25/1M), output pricing ($2.50/1M), context window (1M). No cached input pricing listed for this model. Max output tokens not separately documented.
• https://docs.x.ai/developers/pricing — Confirmed same pricing as models page. Notes that "cached prompt tokens" are a token type but does not list a separate discounted rate for grok-4.20 models.

Fields verified from official sources

Field	Proposed value	Source
input_cost_per_mil_tokens	1.25	xAI developer docs — pricing table
output_cost_per_mil_tokens	2.5	xAI developer docs — pricing table
max_input_tokens	1000000	xAI developer docs — "1M" context window

Fields NOT changed (kept from existing catalog entry)

Field	Existing value	Reason
max_output_tokens	2000000	xAI docs do not publish a separate max output token limit; only "Context: 1M" is shown. Leaving existing value unchanged since it cannot be verified or disproved.
input_cache_read_cost_per_mil_tokens	0.2	xAI docs do not list cached pricing for this model. Not removing since absence from the page does not confirm removal of the feature.
displayName	"Grok 4.20 Reasoning (0309)"	No change needed — already present in catalog.
reasoning	true	No change needed — already present in catalog.
multimodal	true	No change needed — already present in catalog.

sync_models (LiteLLM) cross-check

The LiteLLM model_prices_and_context_window_backup.json catalog does not contain an entry for grok-4.20-0309-reasoning or any xai/grok-4.20-* variant. Therefore no numeric cross-check against sync_models is possible. All proposed values are sourced directly from the official xAI documentation.

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
grok-4.20-0309-reasoning	max_input_tokens	1000000	2000000	xai/grok-4.20-0309-reasoning
grok-4.20-0309-reasoning	input_cost_per_mil_tokens	1.25	2	xai/grok-4.20-0309-reasoning
grok-4.20-0309-reasoning	output_cost_per_mil_tokens	2.5	6	xai/grok-4.20-0309-reasoning

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
ai-proxy	Ready	Preview, Comment	May 20, 2026 9:18am

[chatgpt-codex-connector]

Codex usage limits have been reached for code reviews. Please check with the admins of this repo to increase the limits by adding credits.
Credits must be used to enable repository wide code reviews.