OpenAI · The Reasoning-Model Era
GPT-5
OpenAI's reasoning-first flagship. Native chain-of-thought, three reasoning-effort tiers, the highest published benchmark scores at release.
By C.W. Jameson · Published 19 May 2026 · Last reviewed 19 May 2026
GPT-5 dropped with the model card that explicitly described reasoning effort as a tuneable parameter rather than a model variant. The result is a single endpoint that ranges from sub-second answers to ten-minute deliberations on the same prompt, billed at radically different rates. Operators have to budget for reasoning-tokens or watch invoices climb tenfold.
Field signature
Returns a 'reasoning' field separate from 'content'. Will think before tool calls in a way that's billable.
Specifications
| Released | 2025 |
|---|---|
| Context window | 400,000 tokens |
| Pricing | Variable by reasoning-effort tier |
| Modalities | text · image · audio · video-frames |
| License | Commercial API only |
| Era | The Reasoning-Model Era |
Strengths
- Hardest math and competitive-programming problems
- Multimodal reasoning
Weaknesses
- Pricing surprises
- Refusal rates higher than the 4.x generation
Authentication markers
The fingerprints by which GPT-5 can be identified from its output alone.
| Tell | Meaning |
|---|---|
| Reasoning-effort parameter present in API. | GPT-5 family. |
| Reasoning tokens billed separately from output. | OpenAI o-style billing. |
Notable works
- AIME, IMO, ICPC benchmark records at release
Market position
Reasoning-tier dependent; see OpenAI pricing
Partner offer
OpenAI's API surface remains the broadest commercial offering.
Try OpenAI →Affiliate link — see disclosure.
Primary sources
- [1] OpenAI: Models
From the Almanac shop
The Operator's Compendium
Every agent harness, every routing pattern, every cost trick. 90-page PDF.
$29 — Coming soon