OpenAI model product
GPT-3.5-turbo
5-turbo is an OpenAI model with higher accuracy in responding to requested formats and a fix for a text encoding issue affecting non-English language function calls.
Updated Sep 8, 2026. Default version: GPT-3.5 Turbo
LLMBoard Score0.0GPT-3.5 Turbo
Coverage20%8 benchmark families
Context window16.4KTokens
Official input price$0.50OpenAI API
GPT-3.5-turbo Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-3.5-turbo Benchmark Results
Benchmark scores for GPT-3.5 Turbo.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkDROP | Score70.20% | Rank21 | Participants30 | Percentile31.03% | EvidenceB | Evaluated |
| BenchmarkMGSM | Score56.30% | Rank28 | Participants31 | Percentile10.00% | EvidenceB | Evaluated |
| BenchmarkMathVista | Score0.00% | Rank39 | Participants39 | Percentile0.00% | EvidenceB | Evaluated |
| BenchmarkHumanEval | Score68.00% | Rank58 | Participants66 | Percentile12.31% | EvidenceB | Evaluated |
| BenchmarkMATH | Score43.10% | Rank64 | Participants71 | Percentile10.00% | EvidenceB | Evaluated |
| BenchmarkMMMU | Score0.00% | Rank64 | Participants64 | Percentile0.00% | EvidenceB | Evaluated |
| BenchmarkMMLU | Score69.80% | Rank83 | Participants101 | Percentile18.00% | EvidenceB | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,224.79 rating | Rank205 | Participants210 | Percentile2.39% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text | Score1,124.70 rating | Rank209 | Participants210 | Percentile0.48% | EvidenceA | Evaluated |
| BenchmarkGPQA | Score30.80% | Rank235 | Participants247 | Percentile4.88% | EvidenceB | Evaluated |
GPT-3.5-turbo Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext style control | Categorypolish | Rank153 | Rating / score1,227.34 | Votes144 | ObservationsN/A | Result date |
| Arenatext | Categorypolish | Rank154 | Rating / score1,121.19 | Votes144 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank168 | Rating / score1,142.24 | Votes965 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank169 | Rating / score1,043.38 | Votes965 | ObservationsN/A | Result date |
| Arenatext style control | Categorykorean | Rank171 | Rating / score1,153.04 | Votes1,184 | ObservationsN/A | Result date |
| Arenatext | Categorykorean | Rank172 | Rating / score1,018.45 | Votes1,184 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank176 | Rating / score1,231.10 | Votes1,284 | ObservationsN/A | Result date |
| Arenatext style control | Categoryspanish | Rank176 | Rating / score1,229.62 | Votes895 | ObservationsN/A | Result date |
| Arenatext | Categoryfrench | Rank177 | Rating / score1,117.45 | Votes1,284 | ObservationsN/A | Result date |
| Arenatext | Categoryspanish | Rank178 | Rating / score1,119.66 | Votes895 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank179 | Rating / score1,196.17 | Votes1,885 | ObservationsN/A | Result date |
| Arenatext | Categorygerman | Rank180 | Rating / score1,090.18 | Votes1,885 | ObservationsN/A | Result date |
| Arenatext style control | Categorymath | Rank196 | Rating / score1,204.01 | Votes2,134 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank199 | Rating / score1,274.58 | Votes11,130 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry mathematical | Rank199 | Rating / score1,204.83 | Votes1,746 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank200 | Rating / score1,141.74 | Votes8,626 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank200 | Rating / score1,211.00 | Votes7,558 | ObservationsN/A | Result date |
| Arenatext | Categorychinese | Rank201 | Rating / score1,075.22 | Votes7,558 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry entertainment and sports and media | Rank201 | Rating / score1,195.92 | Votes9,653 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank202 | Rating / score1,128.28 | Votes1,746 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry writing and literature and language | Rank202 | Rating / score1,224.65 | Votes16,149 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexpert | Rank203 | Rating / score1,214.39 | Votes437 | ObservationsN/A | Result date |
| Arenatext style control | Categoryinstruction following | Rank203 | Rating / score1,214.63 | Votes23,523 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry legal and government | Rank204 | Rating / score1,249.15 | Votes3,573 | ObservationsN/A | Result date |
| Arenatext style control | Categoryindustry medicine and healthcare | Rank204 | Rating / score1,227.39 | Votes3,837 | ObservationsN/A | Result date |
| Arenatext style control | Categorymulti turn | Rank204 | Rating / score1,200.58 | Votes10,274 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank205 | Rating / score1,073.53 | Votes3,837 | ObservationsN/A | Result date |
| Arenatext style control | Categorycreative writing | Rank205 | Rating / score1,189.63 | Votes9,694 | ObservationsN/A | Result date |
| Arenatext style control | Categoryexclude ties | Rank205 | Rating / score1,110.51 | Votes44,422 | ObservationsN/A | Result date |
| Arenatext style control | Categorynon english | Rank205 | Rating / score1,208.16 | Votes26,251 | ObservationsN/A | Result date |
GPT-3.5-turbo Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $0.50 input, $1.5 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- From $0.50 input, $1.5 output per 1M via Azure Cognitive Services
- Tracked offerings
- 3
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderOpenAI | Provider model IDgpt-3.5-turbo | Regionglobal | Input / 1M$0.50 | Output / 1M$1.5 | Context16.4K | Updated |
| ProviderAzure Cognitive Services | Provider model IDgpt-3.5-turbo-0125 | Regionglobal | Input / 1M$0.50 | Output / 1M$1.5 | Context16.4K | Updated |
| ProviderAzure | Provider model IDgpt-3.5-turbo-0125 | Regionglobal | Input / 1M$0.50 | Output / 1M$1.5 | Context16.4K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
GPT-3.5-turbo Runtime Performance
Provider-specific output speed and catalog latency for GPT-3.5 Turbo. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderOpenAI | Output Speed100.00 tok/s | Catalog Latency0.50 s | Max Input16.4K | Max Output4.1K | Updated |
| ProviderAzure | Output Speed90.00 tok/s | Catalog Latency0.80 s | Max Input16.4K | Max Output4.1K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-3.5-turbo Specifications
Technical details for the model's default version.
- Version
- GPT-3.5 Turbo
- Released
- Mar 21, 2023
- Knowledge cutoff
- Sep 30, 2021
- Parameters
- N/A
- Context window
- 16.4K
- Max output
- 4.1K
- Inputs
- text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-3.5-turbo Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGPT-3.5 Turbo | Released | LLMBoard0.00 | ParametersN/A | Context16.4K | Max output4.1K | Open weightsNo | LicenseProprietary |
Models similar to GPT-3.5-turbo
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-5.1-Codex-Mini
OpenAI
OP
GPT-4.1-nano
OpenAI
OP
GPT-4
OpenAI
OP
GPT-4o-mini
OpenAI
OP
GPT-4-Turbo
OpenAI
OP
o1 mini
OpenAI
What is GPT-3.5-turbo?
Key information about GPT-3.5-turbo and its available data.
5-turbo is a large language model from OpenAI. It includes improvements to requested-format responses and fixes a text encoding issue for non-English language function calls.
Data as of 2026-09-08.
FAQ
Common questions about GPT-3.5-turbo.
When was GPT-3.5-turbo released?
GPT-3.5-turbo's default version was released on Mar 21, 2023.
How much does GPT-3.5-turbo cost?
GPT-3.5-turbo's official API price is $0.50 per million input tokens and $1.5 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $0.50 input and $1.5 output via Azure Cognitive Services.
Who created GPT-3.5-turbo?
GPT-3.5-turbo was created by OpenAI.
What is the context window for GPT-3.5-turbo?
The default version has a 16.4K token context window.
Is GPT-3.5-turbo open weight?
No. The default version is not marked as having publicly available weights.