GPT-3.5 Turbo (older v0613)
OpenAI · released Jan 25, 2024
- Type
- Proprietary
- Input
- $3.00
- Output
- $4.00
- Cached
- Not published
List price · per 1M tokens · OpenAI at 16K context · source ↗
Our take
Written Aug 3, 2026GPT-3.5 Turbo is an older text-only model from OpenAI, released in early 2024 and still available through a handful of hosts. Its context limit is modest by current standards and it trails newer models on every measured quality dimension, though it remains the cheapest entry point in its own family through certain channels.
Use this only for legacy system maintenance that requires the exact v0613 checkpoint, or when you face extreme budget pressure and can verify the quality suffices through OpenRouter or Azure. Consider it as a low-latency fallback where speed on OpenAI's own servers matters more than output quality. Skip it if you need image or document input, a longer context, or anything approaching current-generation capability.
The case for it
- Cheapest entry in its own family through OpenRouter and Azure, at one-third the input cost and half the output cost of OpenAI direct.
- Fastest throughput on OpenAI's own servers, at more than ten times the speed measured on Azure.
- Coding evaluation is its strongest measured category, roughly fifty points above its overall chat score.
The case against it
- Narrow context window by modern standards, capped at just over four thousand tokens with no larger option.
- Text-only; no image, audio or document input, unlike every current-generation alternative.
- Creative writing is its weakest measured category, more than thirty points below its overall chat score.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)140th of 143 · 1224.7
CodingWriting and fixing code on its own
Arena Coding139th of 143 · 1274.4
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored GPT-3.5 Turbo (older v0613) for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where GPT-3.5 Turbo (older v0613) placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model6 scoresEvery figure we hold, from 6 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Or rent it from someone else
The lab is also the cheapest we hold. The strip above and this offer are the same one, so nothing on this page undercuts OpenAI on 16K of context.
- per 1M tokens
- $3.00 in / $4.00 out
- Context served
- 16K
- Throughput
- ~4 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $1.00 / $2.00 | 4K | not measured | Unknown | Unknown | Unknown |
| Microsoft Azure AI | $1.00 / $2.00 | 4K | 6 tok/s | No | No | Confirmed |
| OpenAI | $3.00 / $4.00 | 16K4K out | 4 tok/s | No | Yesunknown period | Unknown |
OpenRouter
Microsoft Azure AIAcross the 3 listings we hold: 2 say they do not train on prompts, 0 say they do and 1 do not say. 1 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| Microsoft Azure AI | ✓ | ✓ | ✓ |
| OpenAI | ✓ | ✓ | ✓ |
Tool calling: 3 of 3 listings say yes. JSON output: 3 of 3 listings say yes. Strict schema: 3 of 3 listings say yes.
When we formed this view
Dates behind this page
Prices last checked 9d ago
What we do not know about this model yet
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 3 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
Licence and identifiers
What the licence allowsWe hold no licence record for this model. Inside are the identifiers you need to pull it — its Hugging Face repo where we have one, our slug and a machine-readable card.
Licence
Commercial API terms. We hold no licence record for this model, so there is nothing to summarise here.
Identifiers
- Modality record
- text->text
- Catalogue slug
- openai-gpt-3-5-turbo-older-v0613