[BOT ISSUE] xAI: add missing input_cache_read_cost_per_mil_tokens to grok-4.20-0309-non-reasoning

[github-actions]

Gap

The catalog entry for grok-4.20-0309-non-reasoning is missing input_cache_read_cost_per_mil_tokens. The xAI model detail page confirms cached input pricing at $0.20 per million tokens. Every other grok-4.20 variant in the catalog already has this field set to 0.2.

Official sources

Field	Source	URL
Cached input pricing ($0.20/1M)	xAI model detail page	https://docs.x.ai/developers/models/grok-4.20-0309-non-reasoning
Cached pricing availability	xAI models listing page	https://docs.x.ai/developers/models

Verified fields

• Model ID: grok-4.20-0309-non-reasoning — already in catalog at line 11543
• input_cache_read_cost_per_mil_tokens: 0.2 — xAI model detail page explicitly states "Cached input: $0.20" per 1M tokens at the standard tier (under 200k prompt tokens)

Comparison with sibling entries

Model ID	input_cache_read_cost_per_mil_tokens
grok-4.20-0309-reasoning	0.2 ✅
grok-4.20-beta-0309-non-reasoning	0.2 ✅
grok-4.20-beta-0309-reasoning	0.2 ✅
grok-4.20-multi-agent-beta-0309	0.2 ✅
grok-4.20-0309-non-reasoning	missing ❌

The non-reasoning primary entry is the only grok-4.20 variant missing cached pricing.

Verification notes

• Cross-source: Cached pricing confirmed on (1) xAI model detail page (explicit $0.20 value) and (2) xAI models listing page (confirms cached pricing exists across all models)
• Consistency: The child entry grok-4.20-beta-0309-non-reasoning (which has parent: "grok-4.20-0309-non-reasoning") already has this field set to 0.2
• No duplicate issue: No existing open issue references grok-4.20-0309-non-reasoning cached pricing

{
  "kind": "stale_metadata",
  "provider": "xAI",
  "models": ["grok-4.20-0309-non-reasoning"],
  "status": "active",
  "model_specs": {
    "grok-4.20-0309-non-reasoning": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 1.25,
      "output_cost_per_mil_tokens": 2.5,
      "input_cache_read_cost_per_mil_tokens": 0.2,
      "displayName": "Grok 4.20 (Non-Reasoning)",
      "max_input_tokens": 1000000,
      "available_providers": ["xAI"]
    }
  },
  "source_urls": [
    "https://docs.x.ai/developers/models/grok-4.20-0309-non-reasoning",
    "https://docs.x.ai/developers/models"
  ]
}

[github-actions]

Prepared catalog updates for grok-4.20-0309-non-reasoning. Updated models: grok-4.20-0309-non-reasoning.

Verification sources: https://docs.x.ai/developers/models/grok-4.20-0309-non-reasoning, https://docs.x.ai/developers/models
Verified metadata applied:

• grok-4.20-0309-non-reasoning: format=openai, flavor=chat, displayName=Grok 4.20 (Non-Reasoning), max_input_tokens=1000000, max_output_tokens=not provided, pricing(input/output per 1M)=1.25/2.5, input_cache_read_cost_per_mil_tokens=0.2, providers=xAI, multimodal=true

Included in the daily batch PR: #989