omni-moderation-latest
omni-moderation-latest is OpenAI's hosted multimodal moderation alias, not a 2026 safety default we independently crowned. Rechecked 2026-08-17: platform.openai.com/docs/guides/moderation still lists omni-moderation-latest and the dated omni-moderation-2024-09-26 card. The same family still mentions leftover text-moderation-latest. Drop leftover latest safety model / unaudited image+text crown / free as a 2026 census.
Compare Llama Guard 3 if you wanted another leftover safety snapshot (live Meta card is Llama Guard 4 12B), and text-moderation-latest if you wanted the older text-only OpenAI leftover.
Key Features
- Hosted alias, still listed:
omni-moderation-latestremains on the Moderation guide we opened. - Not an independently crowned 2026 default: Recheck that guide before you call it the current OpenAI safety stack.
- Text + image path: The live page still describes omni models as multimodal. Recheck it before you invent extra modalities.
- No invented free/latency row: Do not paste leftover "free forever / millisecond" unless the first-party page still says that on the day you quote it.
Use Cases
- People with an omni-moderation-latest bookmark who need the catalog corrected.
- People comparing leftover text-moderation vs still-listed omni-moderation.
- People who were about to call it OpenAI's 2026 safety default.
Limitation: We did not send sample text or images through the Moderation endpoint.
Pricing
First-party pages on 2026-08-17.
| Piece | Price | Notes |
|---|---|---|
| omni-moderation-latest | Still listed alias | Recheck the Moderation guide. |
| omni-moderation-2024-09-26 | Dated twin | Same family on that page. |
| text-moderation-latest | Leftover text-only | Older path; recast as previous. |
| Dollar / free row | Recheck live docs | Do not invent a 2026 free or per-call census. |
If a leftover blog still calls this the latest OpenAI safety model with no date, treat the dated 2024-09-26 card as the correction.
Getting Started
- Open the Moderation guide before you hard-code
omni-moderation-latest. - Recheck
omni-moderation-2024-09-26if you wanted the dated twin. - Recheck text-moderation-latest if you still had the older text-only leftover.
- Do not paste an unaudited free/latest crown into a deck.
Frequently Asked Questions
Is omni-moderation-latest still listed?
Yes, on the Moderation guide we opened. That does not make it an independently crowned 2026 default.
Same as text-moderation-latest?
No. text-moderation-latest is the older text-only leftover.
Same as Llama Guard 3?
No. Llama Guard 3 is a leftover Meta snapshot. The live Meta card is Llama Guard 4 12B.
Alternatives
- text-moderation-latest: Older OpenAI text-only leftover.
- Llama Guard 3: Leftover Meta safety snapshot; live card is Guard 4 12B.
- GPT-5.6 Sol: Live OpenAI frontier chat if you wandered here for a current model.
Tips
- Call
omni-moderation-latesta still-listed hosted alias, not a 2026 crown. - Recheck the dated
2024-09-26card before you invent a newer snapshot. - Recheck first-party docs before you invent a free or per-call price.
Conclusion
omni-moderation-latest is still listed on OpenAI's Moderation guide, but it is not an independently crowned 2026 safety default. Start at platform.openai.com/docs/guides/moderation, then decide whether the dated omni card or text-moderation-latest already covers the path you need.
Comments
No comments yet. Be the first to comment!
Related Tools
text-moderation-latest
platform.openai.com/docs/guides/moderation
Previous OpenAI text-moderation snapshot. Rechecked: leftover text-only path. Not the 2026 safety default.
OpenAI: tts-1-hd
openai.com/api/pricing
Hosted OpenAI tts-1-hd snapshot. Rechecked: listed at $30/1M. Not the 2026 default TTS path.
OpenAI: tts-1
openai.com/api/pricing
Hosted OpenAI tts-1 snapshot. Rechecked: listed at $15/1M. Not the 2026 default TTS path.
Related Insights

Grok Bot and Hermes Bot: one person finally gets a think tank and a secretariat
Grok Bot now ships with Cursor Pro+. Hermes Bot runs on a VPS. They are not smarter chat boxes. The think tank advises, the secretariat executes, and you still make the call.
Hook OpenCode Go into Codex on Windows. Do Not Open a Second Toolkit.
A ChatGPT-signed Codex desktop app still shows mostly GPT in the picker. On Windows, enable only OpenCode Go and the Grok, GLM, Kimi, DeepSeek, and MiniMax models you already pay for appear in the same selector. Keys stay local. Native GPT stays put.
What Locks Codex Is the Picker, Not the Models
You already pay for OpenCode Go, Grok, and Z.ai, but the Codex picker still shows mostly GPT. The community project codex-router does not teach another install ritual. It puts subscriptions you already bought back into the selector, keeps keys on the machine, and leaves native GPT alone.