LLMIntel
OverviewLive demoCatalogCompareMCPPricingChangelogDocs
Start free
LLMIntel

A verified, normalized feed of AI model lifecycle events across every major provider, so a deprecation shows up on your roadmap instead of in your error logs. Plus runtime telemetry for what those models actually cost you.

Questions? support@llmintel.ai

Product
  • Pricing
  • Compare models
  • Azure OpenAI retirements
  • Dashboard
  • Changelog
  • Contact support
Developers
  • Quickstart
  • Model catalog
  • What’s retiring
  • MCP server
  • API reference
  • OpenAPI spec
  • Lifecycle RSS feed
  • @llmintel/telemetry
Free & open data

The model catalog is a normalized, machine-readable source of truth for AI model lifecycle events. It is free via the public API, and every record links to the primary provider source it came from.

© 2026 Alex Tsimbalistov. All rights reserved.
LLMIntel
OverviewLive demoCatalogCompareMCPPricingChangelogDocs
Start free
← All tracked models

OpenAI model deprecations and shutdown dates

OpenAI deprecates a model first and shuts it down later, usually naming a recommended successor at the same time. Between those two dates the model keeps answering, which is the window to migrate. After shutdown the id returns an error.

We track 51 OpenAI models, 15 of them with a published retirement date. 9 need attention now.

Retiring or deprecated now

Deprecated models still answer but are superseded. Anything inside the 90-day window has a date attached and will stop answering on it.

OpenAI models that are deprecated or retiring soon
ModelStateRetirementReplacement
gpt-4o-audiodeprecated2027-01-20 · in 124 daysopenai/gpt-audio-1.5
gpt-4o-mini-audiodeprecated2027-01-20 · in 124 daysopenai/gpt-audio-1.5
gpt-4o-mini-realtimedeprecated2027-01-20 · in 124 daysopenai/gpt-realtime-2.1-mini
gpt-4o-mini-transcribe-2025-03-20deprecated2027-01-20 · in 124 daysopenai/gpt-4o-mini-transcribe-2025-12-15
gpt-4o-realtimedeprecated2027-01-20 · in 124 daysopenai/gpt-realtime-2.1
gpt-audiodeprecated2027-01-20 · in 124 daysopenai/gpt-audio-1.5
gpt-audio-minideprecated2027-01-20 · in 124 daysopenai/gpt-audio-1.5
gpt-realtimedeprecated2027-01-20 · in 124 daysopenai/gpt-realtime-2.1
gpt-realtime-minideprecated2027-01-20 · in 124 daysopenai/gpt-realtime-2.1-mini

Currently safe

Active, announced, or legacy with no imminent date. Legacy models are superseded but still served.

OpenAI models with no imminent retirement
ModelStateRetirementReplacement
gpt-4.1activeNot announced—
gpt-4.1-2025-04-14activeNot announced—
gpt-4.1-miniactiveNot announced—
gpt-4.1-mini-2025-04-14activeNot announced—
gpt-4.1-nanoactiveNot announced—
gpt-4.1-nano-2025-04-14activeNot announced—
gpt-4oactiveNot announced—
gpt-4o-2024-05-13activeNot announced—
gpt-4o-2024-08-06activeNot announced—
gpt-4o-2024-11-20activeNot announced—
gpt-4o-miniactiveNot announced—
gpt-4o-mini-2024-07-18activeNot announced—
gpt-5activeNot announced—
gpt-5-2025-08-07active2026-12-11 · in 84 days—
gpt-5-miniactiveNot announced—
gpt-5-mini-2025-08-07active2026-12-11 · in 84 days—
gpt-5-nanoactiveNot announced—
gpt-5-nano-2025-08-07active2026-12-11 · in 84 days—
gpt-5-pro-2025-10-06active2026-12-11 · in 84 daysopenai/gpt-5.5-pro
gpt-5.1activeNot announced—
gpt-5.1-2025-11-13activeNot announced—
gpt-5.1-codexactiveNot announced—
gpt-5.1-codex-miniactiveNot announced—
gpt-5.2activeNot announced—
gpt-5.2-2025-12-11activeNot announced—
gpt-5.2-codexactiveNot announced—
gpt-5.3-codexactiveNot announced—
gpt-5.4activeNot announced—
gpt-5.4-2026-03-05activeNot announced—
gpt-5.4-miniactiveNot announced—
gpt-5.4-mini-2026-03-17activeNot announced—
gpt-5.4-nanoactiveNot announced—
gpt-5.4-nano-2026-03-17activeNot announced—
gpt-5.5activeNot announced—
gpt-5.5-2026-04-23activeNot announced—
o3activeNot announced—
o3-2025-04-16active2026-12-11 · in 84 days—
o3-miniactiveNot announced—
o3-mini-2025-01-31activeNot announced—
o3-pro-2025-06-10active2026-12-11 · in 84 daysopenai/gpt-5.5-pro
o4-miniactiveNot announced—
o4-mini-2025-04-16activeNot announced—

Find out which of these you are paying for

A list of dates only helps once you know which of these models your own traffic hits. One line of code meters the models your app actually calls, then reports the monthly spend sitting on each retiring id and what the move to its replacement costs.

See it on sample trafficStart free

Other providers

Azure OpenAIAnthropic ClaudeAWS BedrockGoogle Gemini and Vertex AICohere

Source: OpenAI — Deprecations. Every record links the page it was parsed from and keeps OpenAI’s own wording, so a normalization decision stays checkable.

LLMIntel

A verified, normalized feed of AI model lifecycle events across every major provider, so a deprecation shows up on your roadmap instead of in your error logs. Plus runtime telemetry for what those models actually cost you.

Questions? support@llmintel.ai

Product
  • Pricing
  • Compare models
  • Azure OpenAI retirements
  • Dashboard
  • Changelog
  • Contact support
Developers
  • Quickstart
  • Model catalog
  • What’s retiring
  • MCP server
  • API reference
  • OpenAPI spec
  • Lifecycle RSS feed
  • @llmintel/telemetry
Free & open data

The model catalog is a normalized, machine-readable source of truth for AI model lifecycle events. It is free via the public API, and every record links to the primary provider source it came from.

© 2026 Alex Tsimbalistov. All rights reserved.