Skip to content
Get my AI grade

OpenAI

GPT-5.2-Codex benchmarks, ranks and price

Where Artificial Analysis, LiveBench, LMArena and Berkeley BFCL place GPT-5.2-Codex, what it costs through OpenRouter and Vercel AI Gateway, and the models either side of it on the consensus.

Updated Oct 5, 2026Full leaderboardSources and method

Read Oct 5, 2026

#37of 126

on the LLM leaderboard, with a consensus score of 61 across 2 leaderboards.

Source: Artificial Analysis, LiveBench, LMArena and Berkeley BFCL, combined by Rank.ai.

By leaderboard

Where each leaderboard places GPT-5.2-Codex

Artificial Analysis

#26of 115

Score 28.5. Best run: openai/gpt-5.2-codex.

Intelligence Index: a composite of reasoning, knowledge, maths and coding evaluations.

LiveBench

#17of 36

Score 74.0. Best run: GPT 5.2 Codex.

Contamination-limited questions refreshed each release: reasoning, maths, coding, data analysis, language.

LMArena

Not measured by LMArena.

Blind head-to-head human votes, style controlled.

Berkeley BFCL

Not measured by Berkeley BFCL.

Function calling and tool use, single turn to multi turn and web search.

Price and context

$4.81

OpenRouter: $1.75 in, $14 out. Vercel AI Gateway: $1.75 in, $14 out Context window 400K tokens. LiveBench cost per solved task $0.187.

Nearby

The models either side of GPT-5.2-Codex

  1. #34GLM 4.661
  2. #35Claude Opus 4.161
  3. #36o361
  4. #37GPT-5.2-Codex61
  5. #38DeepSeek V3.2 Exp60
  6. #39GLM 5V Turbo60
  7. #40grok-4-070959

Questions about GPT-5.2-Codex

How good is GPT-5.2-Codex?
GPT-5.2-Codex ranks #37 of 126 models on the Rank.ai LLM leaderboard, with a consensus score of 61 out of 100 across 2 leaderboards.
Is GPT-5.2-Codex better than o3?
Not on the consensus: o3 scores 61 to GPT-5.2-Codex's 61. Compare them source by source below, since the leaderboards test different things.
Is GPT-5.2-Codex better than DeepSeek V3.2 Exp?
On the consensus, yes: 61 to 60.
How much does GPT-5.2-Codex cost?
$1.75 per million input tokens and $14 per million output tokens on OpenRouter, the cheaper listing we read. Blended at three input tokens to one output, that is $4.81 per million.

These models recommend brands. See if they name yours.

Enter your website. In about two minutes, Rank.ai asks ChatGPT, Claude and Gemini 12 questions your customers ask and grades how often they name you.

  • Your grade out of 100How often AI names you, cites your site, and how high it ranks you.
  • Who gets namedEvery competitor in the answers, most named first.
  • The pages AI readsThe sources behind each answer.
  • Three fixesWhat to fix first, with a brief for the first article.