GPT-5.2 Codex
Dec 2025
Retired 2026-07-23Replace withGPT-5.6 SolNewer version
Gemini 4 ArgonCheaper, as goodGPT-6 LunaBetter
MiMo v2.6 ProOpen weights
GPT-5.6 SolNewer versionGemini 4 ArgonCheaper, as goodGPT-6 LunaBetterMiMo v2.6 ProOpen weightsGPT-5.2 CodexCompare inon
Benchmarks of GPT-5.2 Codex22
BenchmarkByScoreRank
Overall1
Knowledge4
Science2
Coding4
Multimodal1
Documents1
Legacy · saturated or no longer reported9
No benchmark matches
Rankingat xhigh effort?
Retired models are left out of the current rankings.
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperClaude Sonnet 5.5max+96%-17%$5.46No forced tool use+5BetterGPT-6 Lunahigh+16%-96%$0.029Function calling only with none · Chat Completions+3OpenMiMo v2.6 Prothinking+63%-89%$0.13Pass back reasoning_content+3CheaperGemini 4 Argonhigh+84%-17%*$1.99Synchronous output stops at 65,536+2CheaperGPT-6.1 Solmax+82%-17%$0.72Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
Claude Sonnet 5.5max+96%-17%$5.46No forced tool use+5Gemini 4 Argonhigh+84%-17%*$1.99Synchronous output stops at 65,536+2GPT-6.1 Solmax+82%-17%$0.72Remove unsupported parameters+5
Muse Spark 1.3max+69%-58%$1.60Only auto tool_choice is supported+5GPT-6 Solmax+67%-17%$1.04ModelScoreCostPer taskWhat changes
GPT-6 Lunahigh+16%-96%$0.029Function calling only with none · Chat Completions+3GPT-5.6 Lunahigh+13%-91%$0.044Claude Haiku 5.5medium+21%-96%$0.047Remove sampling parameters+5MiMo v2.6 FlashthinkingOpen+33%-96%$0.062Pass back reasoning_content+3GPT-6.1 Sollow+48%-17%$0.13Remove unsupported parameters+5ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen+63%-89%$0.13Pass back reasoning_content+3
DeepSeek V4.1 Flash 1017maxOpen+58%-89%$0.26Pass back reasoning_content · Chat Completions+4
GLM 5.3maxOpen+57%-55%$2.01No image input+4
Step 5 PreviewthinkingOpen+53%-70%$1.03web_search not available · Chat Completions+2
Kimi K3maxOpen+53%+25%$2.00Sampling parameters are fixed+3ModelScoreCostPer taskWhat changes
GPT-6 Astraxhigh+84%—$2.31Monitoring can return HTTP 403+2GPT-6.1 Solxhigh+79%—$0.39Remove unsupported parameters+5GPT-6 Solxhigh+55%—$0.52GPT-5.6 Solxhigh+54%—$1.18GPT-5.4xhigh+37%—$2.50 / $15GPT-5.5xhigh+35%—$2.63Prompts with >272K input tokensGPT-5.6 Terraxhigh+33%—$0.63Prompts with >272K input tokensGPT-6 Lunaxhigh+21%—$0.042Function calling only with none · Chat Completions+3All OpenAI models →History
4 versionsBenchmarks
all categoriesChat & Knowledge
GPQA Diamond89.9
Coding
Agents & Tool Use
τ²-bench · AA92.1
Writing & Language
Vision & Documents
Watch out 1
Deprecated›
The official model page marks this model deprecated; plan a migration. Source ↗
FAQ
How much does GPT-5.2 Codex cost?›
$1.75 per 1M input tokens and $14 per 1M output tokens (cheapest host: Azure).
Is GPT-5.2 Codex free to use?›
There is no free API; the cheapest paid price is $1.75 per 1M input tokens.
How can I use GPT-5.2 Codex?›
Through the API of Azure.
What is the context window of GPT-5.2 Codex?›
400K tokens, with up to 128K output tokens.
When was GPT-5.2 Codex released? Is there a newer version?›
Released 2025-12-11. It has been superseded by GPT-5.6 Sol.
Is GPT-5.2 Codex open source?›
No, GPT-5.2 Codex is a closed model available through an API. The closest open-weights option here is MiMo v2.6 Pro.
Can GPT-5.2 Codex read images?›
It accepts text and images as input.
What is GPT-5.2 Codex?›
GPT-5.2 Codex is a closed language model from OpenAI, released Dec 2025.
What are the best alternatives to GPT-5.2 Codex?›
Claude Sonnet 5.5 (cheaper), GPT-6 Luna (better), MiMo v2.6 Pro (open weights).