ResearchLLM Verifications Leaderboard

Data as of · 78 models, 80 questions · reviewed by Suhas Bhairav · methodology

Head to head · tested 27 Oct 2026

GPT-5.3-Codex vs GPT-5.4: accuracy, speed and price

Similar accuracy; GPT-5.3-Codex is faster.

OpenAIGPT-5.3-Codex82.5% score · #3$0.01 per question95 tok/sOpenAIGPT-5.472.5% score · #21$0.011 per question55 tok/s

Verdict

Built from the measurements below

  • Accuracy is too close to call: GPT-5.3-Codex scored 82.5% and GPT-5.4 72.5% on the same 80 questions, a gap within what chance alone could produce.
  • They cost about the same per question ($0.01 vs $0.011).
  • GPT-5.3-Codex generates text 1.7× faster: 95 tokens per second against 55.
  • GPT-5.3-Codex starts answering sooner: 1.4 s to the first token, against 1.7 s for GPT-5.4.
  • GPT-5.4 accepts more context: 1.1M tokens against 400k.

Which should you choose?

Choose GPT-5.3-Codex for the lower cost per correct answer ($0.013 against $0.016), faster streaming and a quicker first token. Choose GPT-5.4 for a 1.1M-token context window. The widest subject gap is in medicine: GPT-5.3-Codex answered 7 of 11 correctly, GPT-5.4 4 of 11. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap +10.0 points (z = 1.51). A winner is called only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.

Head to head

Same questions, same prompt, same API

MeasureGPT-5.3-CodexGPT-5.4Edge
Verification scoreHigher is better82.5%72.5%Too close to call
Cost per questionLower is better$0.01$0.011Similar
Output speedHigher is better9555GPT-5.3-Codex
Time to first tokenLower is better1.4 s1.7 sGPT-5.3-Codex
End-to-end response timeLower is better5.2 s8.7 sGPT-5.3-Codex
Tokens per questionLower is better699706Similar
Context windowHigher is better400k1.1MGPT-5.4

Side by side

Bars are scaled to the larger of the two values. The winner on each measure is marked in bold.

Every measure

GPT-5.3-Codex first, GPT-5.4 second

Verification scorehigher is better
GPT-5.3-Codex82.5%
GPT-5.472.5%
Cost per questionlower is better
GPT-5.3-Codex$0.01
GPT-5.4$0.011
Output speedhigher is better
GPT-5.3-Codex95
GPT-5.455
Time to first tokenlower is better
GPT-5.3-Codex1.4 s
GPT-5.41.7 s
End-to-end response timelower is better
GPT-5.3-Codex5.2 s
GPT-5.48.7 s
Tokens per questionlower is better
GPT-5.3-Codex699
GPT-5.4706
Context windowhigher is better
GPT-5.3-Codex400k
GPT-5.41.1M

Score vs. cost per question

Both models in color, every other model in grey

Most attractive quadrant0255075100$0.00001$0.0001$0.001$0.01$0.10Cost per question (USD, log) →Verification score (%) →Qwen3.8 2.4T A95BCommand AReka Flash 3Gemma 4 26B A4BGemma 4 31BQwen3.5 397B A17BHermes 4 405BMuse WonderGrok 4.7o3gpt-oss-20bMinistral 3 3B 2512Nova Micro 1.0Phi 4Granite 4.2 8BHy3GLM 5.3 FlashSolar Pro 4Trinity Large ThinkingMiniMax M3Mistral Large 4Claude Opus 5.5GPT-6.1 SolGemini 4 ArgonGPT-5.4GPT-6 Lunagpt-oss-120bLlama 4 Maverick

Specifications and pricing

From OpenRouter's catalogue, plus the leaderboard's test dates

SpecificationGPT-5.3-CodexGPT-5.4
CreatorOpenAIOpenAI
Released24 Feb 20265 Mar 2026
Weights——
ReasoningYesYes
Inputtext, image, filetext, image, file
Outputtexttext
Context window400k tokens1.1M tokens
Knowledge cutoff——
List price in / out per 1M$1.75 / $14$2.5 / $15
Blended price per 1M (3:1)$4.81$5.63
Cheapest hostAzure · $1.75 / $14OpenAI · $1.25 / $7.5
Hosts on OpenRouter23
Last tested27 Oct 202625 Oct 2026

Compare any two models →

GPT-5.3-Codex vs GPT-5.4: frequently asked questions

Is GPT-5.3-Codex better than GPT-5.4?
On accuracy they are statistically tied: GPT-5.3-Codex scored 82.5% and GPT-5.4 72.5% on the same 80 questions from open datasets. Similar accuracy; GPT-5.3-Codex is faster.
Which is cheaper, GPT-5.3-Codex or GPT-5.4?
They cost about the same per question as billed: $0.01 for GPT-5.3-Codex and $0.011 for GPT-5.4.
Which is faster, GPT-5.3-Codex or GPT-5.4?
GPT-5.3-Codex is faster at 95 output tokens per second, against 55. Time to first token: 1.4 s for GPT-5.3-Codex, 1.7 s for GPT-5.4.
Which has the larger context window, GPT-5.3-Codex or GPT-5.4?
GPT-5.4, with 1.1M tokens against 400k.