OpenAI: o1-mini
o1-mini is a previous cheap o1 reasoning snapshot, not the 2026 reasoning default. Rechecked 2026-08-18: the o1-mini card still lists it as a small alternative to o1 at $1.10 input / $4.40 output per 1M tokens (cached input $0.55). The same card says use the newer reasoning models instead, and marks o1-mini-2024-09-12 deprecated. The models index now starts at GPT-5.6 Sol. Drop leftover "enhancement of o1 / current cheap reasoning crown."
Compare o1-preview if you wanted the leftover expensive preview twin, GPT-5.6 Sol if you wanted the live start-here model, and GPT-4o if you wanted the leftover omni snapshot.
Key Features
- Still listed at $1.10 / $4.40: Quote the card, not leftover "faster and cheaper" without a number.
- Docs already recommend newer reasoning: The o1-mini page we opened points away from itself.
- Dated twin is deprecated:
o1-mini-2024-09-12. - Not GPT-5.6: Live start is Sol / Terra / Luna.
Use Cases
- People with an o1-mini bookmark who need the catalog corrected.
- People comparing leftover $1.10 o1-mini vs live GPT-5.6 rows.
- People who were about to keep o1-mini as the 2026 cheap reasoning default.
Limitation: We did not re-run leftover coding or math crowns.
Pricing
o1-mini card on 2026-08-18.
| Piece | Price | Notes |
|---|---|---|
| o1-mini | $1.10 / $4.40 per 1M | Cached input $0.55. Still listed. |
| o1-mini-2024-09-12 | Deprecated | Same card. |
| Live start | GPT-5.6 Sol $5 / $30 | gpt-5.6. |
If a leftover blog still calls o1-mini the current cheap reasoning default, treat the models index as the correction.
Getting Started
- Open the o1-mini card and note $1.10 / $4.40.
- Recheck GPT-5.6 Sol if you wanted the live reasoning/coding start.
- Recheck o1-preview if you wanted the leftover $15 / $60 preview.
- Do not paste "current cheap o1" into a 2026 deck.
Frequently Asked Questions
Is o1-mini gone?
No. The card is still listed. The dated snapshot is deprecated, and docs recommend newer reasoning.
Same as o1-preview?
No. Preview leftover copy mixed the two. Preview is the expensive research preview.
Same as GPT-5.6 Luna?
No. Luna is a live cheap GPT-5.6 row. o1-mini is a leftover o1 alias.
Alternatives
- GPT-5.6 Sol: Live OpenAI frontier.
- o1-preview: Leftover expensive o1 preview.
- GPT-4o: Leftover omni snapshot at $2.50 / $10.
Tips
- Quote $1.10 / $4.40 from the live card.
- Call o1-mini previous, not the 2026 reasoning default.
- Recheck GPT-5.6 if you wanted a current start-here model.
Conclusion
o1-mini is a still-listed previous cheap reasoning snapshot at $1.10 / $4.40, not the 2026 default. Start at the o1-mini card, then decide whether GPT-5.6 Sol already covers the path you need.
Comments
No comments yet. Be the first to comment!
Related Tools
OpenAI: o1-preview
developers.openai.com/api/docs/models/o1-preview
Previous o1-preview snapshot. Rechecked: still $15/$60; dated twin deprecated. Not a cheaper o1 and not 2026 default.
DeepSeek-R1
www.deepseek.com
Previous DeepSeek-R1 snapshot. Rechecked: live path is V4 Flash/Pro. Not an o1 / 2026 reasoning default.
Gemini 2.0 Flash Thinking
ai.google.dev/gemini-api/docs/models/gemini-2.0-flash
Previous Gemini 2.0 Flash Thinking snapshot. Rechecked: 2.0 Flash family shut down June 1, 2026. Not a 2026 o1-mini alternative.
Related Insights
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.
Seven AI Coding CLIs, Six Months: No Matter How Strong the Model, Work Needs Supervision
Claude Code, Codex, opencode, pi, omp and DeepSeek Harness all have personalities. After six months of deep use I run a division of labor: pi for the fastest cheapest reviews, omp for complex PRs, DeepSeek Harness on V4 Flash for high-frequency low-cost review, and Claude Code, Qoder and Cursor for writing. No matter how strong the model, work needs supervision — ideally from an independent third party.