text model
GPT-3.5 Turbo
OpenAIThe model that started it all — and the cost floor everything is measured against.
GPT-3.5 Turbo is the baseline of this catalogue. It is genuinely cheap and genuinely fast, and for narrow classification it is still occasionally the right tool. For anything requiring judgement it is not competitive, and OpenAI has retired it from new deployments.
CapabilitiesTool useJSON mode
Variants
2 ways to run it.
Same family, different quality/cost rung. Picking the wrong one is the most common way teams overpay.
Turbo (16K)
Balancedgpt-3.5-turbo-0125Final and best 3.5 release.
1×
rel. cost
34
quality
Instruct
Fastgpt-3.5-turbo-instructCompletion-style, not chat.
3×
rel. cost
28
quality
Fit
Two audiences, one model.
The same model is a different proposition depending on what you point it at.
22/100
Recruitment fit
Do not screen candidates with this. Its judgement is not good enough to put in front of a hiring decision.
Best for
- Nothing candidate-facing
Strengths
- Very cheap
- Very fast
Watch out for
- 16K context cannot hold a full CV plus a job spec plus a rubric
- Materially higher error rate on structured extraction
The numbers
Measured, not asserted.
Published benchmarks
Vendor and independent figures.
- MMLU
- 70%Published