[BOT ISSUE] Groq: missing deprecation flags for meta-llama/llama-4-maverick-17b-128e-instruct and moonshotai/kimi-k2-instruct-0905

[github-actions]

Summary

Two Groq-only models are past their Groq shutdown dates but have no deprecated or deprecation_date flags in the catalog. Both still appear as active, available models.

Model ID	Groq Shutdown Date	Days Since Shutdown	Catalog Status
meta-llama/llama-4-maverick-17b-128e-instruct	2026-03-09	107	Active (no deprecation flag)
moonshotai/kimi-k2-instruct-0905	2026-04-15	70	Active (no deprecation flag)

Both models have available_providers: ["groq"] — Groq is their only provider. With Groq having shut them down, these are dead endpoints.

Sources

• https://console.groq.com/docs/deprecations (lists both models with explicit shutdown dates)
• https://groq.com/pricing/ (neither model appears in current active pricing)

Verification

• Both models are Groq-only (available_providers: ["groq"]), so no other provider serves them.
• Shutdown dates confirmed on Groq deprecations page; absence confirmed on pricing page (two independent signals).
• Groq recommends openai/gpt-oss-120b as successor for both.
• No existing open issue covers these deprecations.

Metadata

{
  "kind": "stale_metadata",
  "provider": "groq",
  "model_specs": {
    "meta-llama/llama-4-maverick-17b-128e-instruct": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 0.2,
      "output_cost_per_mil_tokens": 0.6,
      "displayName": "Llama 4 Maverick (17Bx128E)",
      "experimental": true,
      "deprecated": true,
      "deprecation_date": "2026-03-09",
      "max_input_tokens": 131072,
      "max_output_tokens": 8192,
      "available_providers": ["groq"]
    },
    "moonshotai/kimi-k2-instruct-0905": {
      "format": "openai",
      "flavor": "chat",
      "input_cost_per_mil_tokens": 1,
      "output_cost_per_mil_tokens": 3,
      "input_cache_read_cost_per_mil_tokens": 0.5,
      "displayName": "Kimi K2 0905",
      "experimental": true,
      "deprecated": true,
      "deprecation_date": "2026-04-15",
      "max_input_tokens": 262144,
      "max_output_tokens": 16384,
      "available_providers": ["groq"]
    }
  }
}

[github-actions]

Prepared catalog updates for meta-llama/llama-4-maverick-17b-128e-instruct, moonshotai/kimi-k2-instruct-0905. Updated models: meta-llama/llama-4-maverick-17b-128e-instruct, moonshotai/kimi-k2-instruct-0905.

Verification sources: https://console.groq.com/docs/deprecations, https://groq.com/pricing/
Verified metadata applied:

• meta-llama/llama-4-maverick-17b-128e-instruct: format=openai, flavor=chat, displayName=Llama 4 Maverick (17Bx128E), max_input_tokens=131072, max_output_tokens=8192, pricing(input/output per 1M)=0.2/0.6, providers=groq, multimodal=true, deprecated=true, deprecation_date=2026-03-09
• moonshotai/kimi-k2-instruct-0905: format=openai, flavor=chat, displayName=Kimi K2 0905, max_input_tokens=262144, max_output_tokens=16384, pricing(input/output per 1M)=1/3, input_cache_read_cost_per_mil_tokens=0.5, providers=groq, deprecated=true, deprecation_date=2026-04-15

Included in the daily batch PR: #876