text model
GPT-6
OpenAIFrontier reasoning with an effort dial — from instant to deliberate.
GPT-6 is the current frontier of the GPT line. The headline change over GPT-5 is not raw benchmark score but controllability: a single effort parameter moves the same model from sub-second replies to multi-minute deliberate reasoning, so one integration covers both a live chat widget and an overnight batch job. For hiring teams that matters because screening and structured interview scoring have opposite latency budgets, and you no longer need two models to serve them.
Announced model. All pricing, benchmarks and evals are estimates — replace before this page is indexed.
Variants
5 ways to run it.
Same family, different quality/cost rung. Picking the wrong one is the most common way teams overpay.
Max effort
Maxgpt-6-maxLongest deliberation. Reserve it for work where a wrong answer is expensive — final scoring, offer-risk analysis, contract review.
Final interview scoring · Complex multi-step analysis
High
Highgpt-6-highThe default for anything a human will act on. Strong multi-step reasoning without the max-effort latency tax.
Structured interview scoring · Long-document analysis
Medium
Balancedgpt-6-mediumThe workhorse. Good judgement at a price that survives volume — this is the rung most production traffic should sit on.
CV parsing at volume · Drafting and summarisation
Low
Fastgpt-6-lowNear-instant. Skips deliberation entirely — right for classification, routing and anything a user is waiting on.
Live chat · Intent routing · Tagging
Mini
Minigpt-6-miniSmaller sibling. The cheapest way to run GPT-6-shaped prompts when the task is mechanical rather than judgemental.
Bulk extraction · Deduplication · Cheap pre-filters
Fit
Two audiences, one model.
The same model is a different proposition depending on what you point it at.
Recruitment fit
The strongest general model for hiring work — but only worth its price on the judgement-heavy steps, not on parsing.
Best for
- Structured interview scoring against a rubric
- Comparing a shortlist against a job spec
- Drafting candidate feedback that a human will edit
Strengths
- Holds a full interview transcript and a job spec in context without chunking
- Effort dial lets live screening and overnight scoring share one integration
- Structured outputs are reliable enough to write straight into an ATS record
Watch out for
- Max effort is roughly 24× the cost of Mini — do not point bulk CV parsing at it
- A frontier model is not a fairness guarantee: you still need audited rubrics and a human on the shortlist
- Pricing is unpublished, so total cost of ownership is unmodellable today
The numbers
Measured, not asserted.
HumanLike evals
Run on our own harness.
- CV field extraction (F1)
- 0.94Estimate
- Time to first token (medium effort)
- 640 msEstimate
Projection. Not yet run against GPT-6.
Published benchmarks
Vendor and independent figures.
- MMLU-Pro
- TBDEstimate
- SWE-bench Verified
- TBDEstimate
