[BOT ISSUE] Groq: missing deprecation flags for llama3-8b-8192, mistral-saba-24b, and deepseek-r1-distill-llama-70b

[github-actions]

Summary

Three models served by Groq are past their official shutdown dates but have no deprecated flag in the catalog. These are dead endpoints that will fail on any API call.

Model ID	Shutdown Date	Days Since Shutdown	Other Providers
llama3-8b-8192	2025-08-30	302	Groq only
mistral-saba-24b	2025-07-30	333	Groq only
deepseek-r1-distill-llama-70b	2025-10-02	268	Groq + Cerebras (also deprecated on Cerebras 2025-08-12)

deepseek-r1-distill-llama-70b is deprecated on both of its providers (Groq shutdown 2025-10-02, Cerebras shutdown 2025-08-12) — no live endpoint remains.

Sources

• Groq deprecations page — lists all three with explicit shutdown dates
• Groq pricing page — none of these models are listed (confirming shutdown)
• Cerebras deprecation page — confirms deepseek-r1-distill-llama-70b deprecated 2025-08-12 on Cerebras

Verification

Check	Status	Detail
Cross-source	✅	Shutdown dates confirmed on deprecations page; absence confirmed on pricing page
Already fixed	✅	Confirmed no deprecation flags on any of these entries in model_list.json
ID format	✅	All IDs match existing catalog entries
Duplicate check	✅	Not covered by any open issue

Local Files Inspected

• packages/proxy/schema/model_list.json — llama3-8b-8192 at line ~5493, mistral-saba-24b at line ~5505, deepseek-r1-distill-llama-70b at line ~3641

{
  "kind": "stale_metadata",
  "provider": "groq",
  "models": ["llama3-8b-8192", "mistral-saba-24b", "deepseek-r1-distill-llama-70b"],
  "status": "active",
  "model_specs": {
    "llama3-8b-8192": {
      "format": "openai",
      "flavor": "chat",
      "input_cost_per_mil_tokens": 0.05,
      "output_cost_per_mil_tokens": 0.08,
      "displayName": "Llama 3 8B 8k",
      "deprecated": true,
      "deprecation_date": "2025-08-30",
      "available_providers": ["groq"]
    },
    "mistral-saba-24b": {
      "format": "openai",
      "flavor": "chat",
      "input_cost_per_mil_tokens": 0.79,
      "output_cost_per_mil_tokens": 0.79,
      "displayName": "Mistral Saba 24B",
      "deprecated": true,
      "deprecation_date": "2025-07-30",
      "available_providers": ["groq"]
    },
    "deepseek-r1-distill-llama-70b": {
      "format": "openai",
      "flavor": "chat",
      "input_cost_per_mil_tokens": 0.23,
      "output_cost_per_mil_tokens": 0.69,
      "displayName": "DeepSeek R1 Distill Llama 70b",
      "experimental": true,
      "deprecated": true,
      "deprecation_date": "2025-08-12",
      "available_providers": ["groq", "cerebras"]
    }
  },
  "source_urls": [
    "https://console.groq.com/docs/deprecations",
    "https://groq.com/pricing/",
    "https://inference-docs.cerebras.ai/support/deprecation"
  ]
}

[github-actions]

Prepared catalog updates for llama3-8b-8192, mistral-saba-24b, deepseek-r1-distill-llama-70b. Updated models: llama3-8b-8192, mistral-saba-24b, deepseek-r1-distill-llama-70b.

Verification sources: https://console.groq.com/docs/deprecations, https://groq.com/pricing/, https://inference-docs.cerebras.ai/support/deprecation
Verified metadata applied:

• llama3-8b-8192: format=openai, flavor=chat, displayName=Llama 3 8B 8k, max_input_tokens=not provided, max_output_tokens=not provided, pricing(input/output per 1M)=0.05/0.08, providers=groq, deprecated=true, deprecation_date=2025-08-30
• mistral-saba-24b: format=openai, flavor=chat, displayName=Mistral Saba 24B, max_input_tokens=not provided, max_output_tokens=not provided, pricing(input/output per 1M)=0.79/0.79, providers=groq, deprecated=true, deprecation_date=2025-07-30
• deepseek-r1-distill-llama-70b: format=openai, flavor=chat, displayName=DeepSeek R1 Distill Llama 70b, max_input_tokens=not provided, max_output_tokens=not provided, pricing(input/output per 1M)=0.23/0.69, providers=groq|cerebras, deprecated=true, deprecation_date=2025-08-12

Included in the daily batch PR: #876