OpenAI model product
GPT-5.2
2 is an OpenAI large language model available in Instant, Thinking, and Pro variants, with support for up to 256k tokens of context and tool-calling.
Updated Oct 23, 2026. Default version: GPT-5.2
LLMBoard Score68.0GPT-5.2
Coverage60%23 benchmark families
Context window400KTokens
Official input price$1.75OpenAI API
GPT-5.2 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
GPT-5.2 Benchmark Results
Benchmark scores for GPT-5.2.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkBrowseComp Long Context 128k | Score92.00% | Rank01 | Participants5 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkBrowseComp Long Context 256k | Score89.80% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkAIME 2025 | Score100.00% | Rank02 | Participants122 | Percentile99.17% | EvidenceC | Evaluated |
| BenchmarkGraphwalks BFS <128k | Score94.00% | Rank02 | Participants13 | Percentile91.67% | EvidenceC | Evaluated |
| BenchmarkHMMT 2025 | Score99.40% | Rank02 | Participants33 | Percentile96.88% | EvidenceC | Evaluated |
| BenchmarkGraphwalks parents <128k | Score89.00% | Rank03 | Participants12 | Percentile81.82% | EvidenceC | Evaluated |
| BenchmarkScreenSpot Pro | Score86.30% | Rank03 | Participants26 | Percentile92.00% | EvidenceC | Evaluated |
| BenchmarkSWE-Lancer (IC-Diamond subset) | Score74.60% | Rank03 | Participants6 | Percentile60.00% | EvidenceC | Evaluated |
| BenchmarkTau2 Telecom | Score98.70% | Rank04 | Participants37 | Percentile91.67% | EvidenceC | Evaluated |
| BenchmarkVideoMMMU | Score85.90% | Rank04 | Participants28 | Percentile88.89% | EvidenceC | Evaluated |
| BenchmarkFrontierMath | Score40.30% | Rank05 | Participants17 | Percentile75.00% | EvidenceC | Evaluated |
| BenchmarkARC-AGI | Score86.20% | Rank06 | Participants12 | Percentile54.55% | EvidenceC | Evaluated |
| BenchmarkLM Arena Search Factuality | Score1,224.01 rating | Rank06 | Participants33 | Percentile84.38% | EvidenceA | Evaluated |
| BenchmarkTau2 Retail | Score82.00% | Rank07 | Participants28 | Percentile77.78% | EvidenceC | Evaluated |
| BenchmarkARC-AGI v2 | Score52.90% | Rank09 | Participants19 | Percentile55.56% | EvidenceC | Evaluated |
| BenchmarkMMMLU | Score89.60% | Rank09 | Participants51 | Percentile84.00% | EvidenceC | Evaluated |
| BenchmarkLiveBench math (2026-06-25) | Score93.17 score | Rank12 | Participants43 | Percentile73.81% | EvidenceB | EvaluatedN/A |
| BenchmarkGPQA | Score92.40% | Rank14 | Participants256 | Percentile94.90% | EvidenceC | Evaluated |
| BenchmarkLM Arena Search | Score1,207.24 rating | Rank14 | Participants33 | Percentile59.38% | EvidenceA | Evaluated |
| BenchmarkLiveBench | Score74.84% | Rank16 | Participants38 | Percentile59.46% | EvidenceB | Evaluated |
| BenchmarkMMMU-Pro | Score79.50% | Rank16 | Participants73 | Percentile79.17% | EvidenceC | Evaluated |
| BenchmarkSWE-Bench Verified | Score80.00% | Rank17 | Participants120 | Percentile86.55% | EvidenceC | Evaluated |
| BenchmarkCharXiv-R | Score82.10% | Rank21 | Participants60 | Percentile66.10% | EvidenceC | Evaluated |
| BenchmarkLiveBench reasoning (2026-06-25) | Score83.21 score | Rank26 | Participants43 | Percentile40.48% | EvidenceB | EvaluatedN/A |
| BenchmarkLM Arena Search Style Control | Score1,161.40 rating | Rank27 | Participants33 | Percentile18.75% | EvidenceA | Evaluated |
| BenchmarkToolathlon | Score46.30% | Rank29 | Participants42 | Percentile31.71% | EvidenceC | Evaluated |
| BenchmarkMCP Atlas | Score60.60% | Rank34 | Participants39 | Percentile13.16% | EvidenceC | Evaluated |
| BenchmarkLM Arena Document Style Control | Score1,425.37 rating | Rank35 | Participants38 | Percentile8.11% | EvidenceA | Evaluated |
| BenchmarkLM Arena Document | Score1,404.59 rating | Rank36 | Participants38 | Percentile5.41% | EvidenceA | Evaluated |
| BenchmarkLiveBench instruction (2026-06-25) | Score61.77 score | Rank36 | Participants43 | Percentile16.67% | EvidenceB | EvaluatedN/A |
GPT-5.2 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenavision style control | Categorycaptioning | Rank03 | Rating / score1,262.12 | Votes94 | ObservationsN/A | Result date |
| Arenasearch factuality | Categoryoverall | Rank06 | Rating / score1,224.01 | Votes55,464 | ObservationsN/A | Result date |
| Arenavision | Categorycaptioning | Rank08 | Rating / score1,266.64 | Votes94 | ObservationsN/A | Result date |
| Arenasearch | Categoryoverall | Rank14 | Rating / score1,207.24 | Votes55,634 | ObservationsN/A | Result date |
| Arenatext factuality | Categorypolish | Rank17 | Rating / score1,441.17 | Votes1,095 | ObservationsN/A | Result date |
| Arenatext factuality | Categorygerman | Rank19 | Rating / score1,421.68 | Votes726 | ObservationsN/A | Result date |
| Arenavision | Categoryentity recognition | Rank20 | Rating / score1,231.96 | Votes168 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryfrench | Rank23 | Rating / score1,489.45 | Votes976 | ObservationsN/A | Result date |
| Arenavision style control | Categoryentity recognition | Rank23 | Rating / score1,201.30 | Votes168 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryjapanese | Rank24 | Rating / score1,418.23 | Votes510 | ObservationsN/A | Result date |
| Arenasearch style control | Categoryoverall | Rank27 | Rating / score1,161.40 | Votes55,634 | ObservationsN/A | Result date |
| Arenatext factuality | Categorykorean | Rank31 | Rating / score1,400.40 | Votes742 | ObservationsN/A | Result date |
| Arenavision style control | Categorychinese | Rank34 | Rating / score1,299.77 | Votes1,351 | ObservationsN/A | Result date |
| Arenadocument style control | Categoryoverall | Rank35 | Rating / score1,425.37 | Votes28,212 | ObservationsN/A | Result date |
| Arenadocument | Categoryoverall | Rank36 | Rating / score1,404.59 | Votes7,108 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymath | Rank38 | Rating / score1,454.60 | Votes2,044 | ObservationsN/A | Result date |
| Arenatext style control | Categoryjapanese | Rank39 | Rating / score1,436.16 | Votes446 | ObservationsN/A | Result date |
| Arenavision | Categorychinese | Rank39 | Rating / score1,304.84 | Votes1,351 | ObservationsN/A | Result date |
| Arenatext | Categoryjapanese | Rank40 | Rating / score1,419.69 | Votes446 | ObservationsN/A | Result date |
| Arenatext style control | Categorygerman | Rank41 | Rating / score1,463.63 | Votes834 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry legal and government | Rank42 | Rating / score1,467.05 | Votes2,468 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhomework | Rank42 | Rating / score1,281.69 | Votes2,888 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry mathematical | Rank43 | Rating / score1,453.85 | Votes1,738 | ObservationsN/A | Result date |
| Arenavision style control | Categorydiagram | Rank43 | Rating / score1,275.99 | Votes4,773 | ObservationsN/A | Result date |
| Arenavision style control | Categoryocr | Rank46 | Rating / score1,260.05 | Votes12,905 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry medicine and healthcare | Rank47 | Rating / score1,473.73 | Votes4,473 | ObservationsN/A | Result date |
| Arenatext style control | Categoryfrench | Rank47 | Rating / score1,479.80 | Votes1,439 | ObservationsN/A | Result date |
| Arenavision | Categoryhomework | Rank47 | Rating / score1,276.24 | Votes2,888 | ObservationsN/A | Result date |
| Arenavision | Categoryocr | Rank48 | Rating / score1,256.91 | Votes12,905 | ObservationsN/A | Result date |
| Arenavision style control | Categoryhumor | Rank48 | Rating / score1,217.39 | Votes681 | ObservationsN/A | Result date |
GPT-5.2 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $1.75 input, $14 output per 1M
- Official provider
- OpenAI
- Lowest third-party
- From $1.59 input, $12.73 output per 1M via Poe
- Tracked offerings
- 21
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderPoe | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.59 | Output / 1M$12.73 | Context400K | Updated |
| ProviderOrcaRouter | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderOrcaRouter | Provider model IDopenai/gpt-5.2-chat-latest | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context128K | Updated |
| ProviderNanoGPT | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderImpossibl | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderOpenRouter | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderOpenRouter | Provider model IDopenai/gpt-5.2-chat | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context128K | Updated |
| ProviderOpenAI | Provider model IDgpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderLLM Gateway | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderNEAR AI Cloud | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderMerge Gateway | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderMerge Gateway | Provider model IDopenai/gpt-5.2-chat-latest | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context128K | Updated |
| ProviderZenMux | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderPerplexity Agent | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderOfox | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderVercel AI Gateway | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderEden AI | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderKilo Gateway | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
| ProviderPrime Inference | Provider model IDopenai/gpt-5.2 | Regionglobal | Input / 1M$1.75 | Output / 1M$14 | Context400K | Updated |
GPT-5.2 Runtime Performance
Provider-specific output speed and catalog latency for GPT-5.2. 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 Latency2.00 s | Max Input400K | Max Output128K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
GPT-5.2 Specifications
Technical details for the model's default version.
- Version
- GPT-5.2
- Released
- Dec 11, 2025
- Knowledge cutoff
- Aug 25, 2025
- Parameters
- N/A
- Context window
- 400K
- Max output
- 128K
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
GPT-5.2 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-5.2 | Released | LLMBoard68.00 | ParametersN/A | Context400K | Max output128K | Open weightsNo | LicenseProprietary |
Models similar to GPT-5.2
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
OP
GPT-6-Luna
OpenAI
OP
GPT-5.2-Pro
OpenAI
OP
GPT-5.6-Luna
OpenAI
OP
GPT-5.4
OpenAI
OP
GPT-5.3-Codex
OpenAI
OP
GPT-5.5-Pro
OpenAI
What is GPT-5.2?
Key information about GPT-5.2 and its available data.
2 is an OpenAI large language model for professional knowledge work, coding, science, mathematics, long-context tasks, and tool-calling. 75/1M input tokens and $14/1M output tokens, with Pro variants supporting xhigh reasoning.
Data as of 2026-10-22.
FAQ
Common questions about GPT-5.2.
When was GPT-5.2 released?
GPT-5.2's default version was released on Dec 11, 2025.
How much does GPT-5.2 cost?
GPT-5.2's official API price is $1.75 per million input tokens and $14 per million output tokens via OpenAI. The lowest tracked third-party offer starts at $1.59 input and $12.73 output via Poe.
Who created GPT-5.2?
GPT-5.2 was created by OpenAI.
What is the context window for GPT-5.2?
The default version has a 400K token context window.
Is GPT-5.2 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer GPT-5.2?
21 provider offerings are linked to the default version.