Use this MCP server to check and track LLM model lifecycle status across major providers so agents avoid deprecated or retired model IDs and find replacements.
Check whether a specific model ID/alias is active, deprecated, retiring, or retired, including days to/after retirement, pricing/limits, source link, and replacement guidance.
List models retiring within a configurable window, default 180 days and up to 1095 days, soonest first and filterable by provider.
Suggest provider-recommended replacements for retiring/deprecated models, or same-provider active candidates when no successor is published.
Search the model catalog by provider and lifecycle state, with a configurable result limit.
Review recent lifecycle changes across all providers, default past 30 days and up to 365 days, filterable by provider.
Query OpenAI, Anthropic, Azure, AWS Bedrock, Google, Cohere, and Mistral lifecycle data without an API key via MCP host or hosted HTTP endpoint.
This server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
An MCP server that tells your coding agent whether a model id is safe to use.
LLMs are trained on a snapshot of the world and will confidently write gpt-4-32k into your code
long after it stops answering. This server gives the agent a live lookup for whether a model is
deprecated and when it stops working. It returns the replacement too. Answers are normalized across
OpenAI, Anthropic, Azure AI Foundry, AWS Bedrock, Google, and Cohere, and parsed from each
provider's own deprecation pages.
No API key, no signup. The catalog is public.
Add it to any MCP host. The package runs straight from npm via npx.
In .cursor/mcp.json:
{
"mcpServers": {
"llmintel": {
"command": "npx",
"args": ["-y", "@llmintel/mcp"]
}
}
}claude mcp add llmintel -- npx -y @llmintel/mcpSame shape as the Cursor block above, in claude_desktop_config.json.
The same five tools are served over Streamable HTTP at https://llmintel.ai/v1/mcp. Hosts that take
a URL need no Node and no package:
{
"mcpServers": {
"llmintel": {
"url": "https://llmintel.ai/v1/mcp"
}
}
}The endpoint is stateless and read-only. It answers from the same catalog the npm package queries.
Related MCP server: endoflife-mcp
Tool | Returns |
| Whether one model id is safe to use: lifecycle state, the retirement deadline in days, the replacement, and the source link. |
| What breaks in the next 180 days (default; configurable up to 1095). Past-due models are listed first, then upcoming ones soonest-first. |
| The provider's own recommendation for what to move to. Falls back to same-provider active models when none was published. |
| Catalog search filtered by provider and lifecycle state. |
| The change feed across all providers, for questions like "what was deprecated this month". |
You: Before we ship this, check the model ids in
src/agents/.
The agent calls check_model for each one and gets back:
DO NOT USE — this model is retired; API calls to it fail.
"claude-sonnet-4-20250514" resolves to the tracked model anthropic/claude-sonnet-4-20250514.
Model: claude-sonnet-4-20250514 (anthropic/claude-sonnet-4-20250514)
Provider: anthropic
Lifecycle state: retired — retired; calls fail
Deprecated: 2026-04-14 (105 days ago)
Retirement: 2026-06-15 (43 days ago)
The provider has not named a replacement. Use suggest_replacement for options.
Pricing/limits: $3/1M in · $15/1M out
Source: https://docs.anthropic.com/en/docs/about-claude/model-deprecations
Provider's own term: "Retired"Deadlines are always given in days, because a model cannot reliably judge whether 2026-07-30 is
soon.
A failed lookup is never a safety verdict. If the catalog is unreachable, the tool returns an MCP error and says so. An agent that read a network failure as "no deprecation found" would happily ship a retired model id. A model that simply isn't tracked gets the same treatment: it returns "not in the catalog, verify with the provider", never "OK".
Pass whatever string is literally in the code (gpt-4o, anthropic/claude-opus-4-1, azure/gpt-4o)
and it resolves to the canonical tracked model.
When the provider's own deprecation notice names a successor, that is what you get. Otherwise the fallback list of same-provider active models is labelled as candidates to evaluate, so an agent can tell the two apart.
Anything past its retirement date is broken now, so it gets its own heading instead of sitting in "retiring soon".
Both variables are optional.
Variable | Default | Purpose |
| none | Raises the rate-limit budget. The catalog itself is public, so you do not need this. |
|
| Point at a self-hosted or staging catalog. |
Anonymous callers get 30 requests/minute per IP, enough for interactive agent use.
Every record links to the provider page it was parsed from and preserves the provider's verbatim
lifecycle term (sourceTerm), so a normalization decision is always auditable. Changes go through a
human verification queue before publication. Collector freshness is public at
/v1/status.
The same data is available as a plain REST API, also without a key. See llmintel.ai/docs.
Directory | Entry |
Official MCP Registry |
|
| |
| |
|
pnpm --filter @llmintel/mcp build
pnpm exec vitest run packages/mcp # protocol-level tests against a fake catalog
# Drive the built binary against a live catalog
pnpm --filter @llmintel/mcp smoke
LLMINTEL_BASE_URL=http://localhost:3000 pnpm --filter @llmintel/mcp smokeMIT © LLMIntel
Check whether an LLM model id is safe to use, deprecated, or retired, and what to migrate to. Accepts the exact string used in code (e.g. 'gpt-4o', 'claude-sonnet-4-5-20250929', 'anthropic/claude-opus-4-1'). Call this before writing or changing any hardcoded model id.
Inputs: model (required).
List tracked models scheduled to retire within a time window, soonest first. Use this to audit a codebase or plan migration work.
Inputs: provider, withinDays.
Lifecycle state changes across all tracked providers, newest first. Use this to answer 'what model deprecations happened recently?'.
Inputs: provider, sinceDays.
List tracked models, optionally filtered by provider and lifecycle state. Use this to discover what is currently available from a provider.
Inputs: limit, state, provider.
Given a deprecated or retiring model, return the provider's recommended replacement(s) and, when none is published, active models from the same provider to consider.
Inputs: model (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 5 tools
Each tool serves a distinct purpose: recent changes, model check, retiring list, replacement suggestion, and filtered search. No overlap or confusion.
Four tools follow verb_noun snake_case pattern (check_model, list_retiring_models, suggest_replacement, search_models), but recent_lifecycle_changes uses an adjective_noun pattern, creating a minor inconsistency.
Five tools are well-suited for a model lifecycle server, covering essential queries without being excessive or sparse.
The tool surface covers key lifecycle needs: checking status, listing retiring models, suggesting replacements, viewing recent changes, and search. No obvious gaps for a read-only information server.
Check if an AI model is deprecated, retiring, or silently changed price or context window.
Find retiring AI models and outdated API SDKs in your code, with dates, replacements and guides.
AI model shutdown dates, replacements, and migration diffs. Free, no API key.
AI model shutdown dates, replacements, change feed and webhook alerts for 8 providers.
