GPT-5.3-Codex vs GPT-5.4 Pro
Full profiles: GPT-5.3-Codex (OpenAI) and GPT-5.4 Pro (OpenAI): every benchmark, price and rank.
Updated · Data from 7 sources
As of 2 Nov 2026, GPT-5.3-Codex and GPT-5.4 Pro are neck and neck on quality (62.9 vs 63.1 on the BenchLeader Index, inside the ±7.8 margin), and GPT-5.3-Codex is 14× cheaper at $4.81 vs $67.50 per 1M tokens blended. Choose GPT-5.3-Codex for instruction following (74 vs 63), where it also streams 3×+ faster; choose GPT-5.4 Pro for fast first answers (8.2 s vs 64.7 s) and multimodal (70 vs 61).
Key facts (as of 2 Nov 2026)
- GPT-5.3-Codex scores 62.9 on the BenchLeader Index and GPT-5.4 Pro scores 63.1 (BenchLeader, 2 Nov 2026).
- GPT-5.3-Codex costs $4.81 per 1M tokens blended (list price, 2 Nov 2026).
- GPT-5.4 Pro costs $67.50 per 1M tokens blended (list price, 2 Nov 2026).
- GPT-5.3-Codex streams 3×+ faster than GPT-5.4 Pro (Artificial Analysis, 7-day median, 2 Nov 2026).
GPT-5.3-Codex benchmarks Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Images and other media | GPT-5.4 Pro | 70 vs 61 category score | High |
| Following instructions | GPT-5.3-Codex | 74 vs 63 category score | High |
| Lowest cost | GPT-5.3-Codex | $4.81 vs $67.50 per 1M blended; agentic coding session $0.161 vs $2.52 | – |
| Fastest first answer | GPT-5.4 Pro | 8.2 s (GPT-5.4 Pro) vs 64.7 s (GPT-5.3-Codex (xhigh)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- GPT-5.3-Codex (xhigh) and GPT-5.4 Pro are neck and neck on quality (62.9 vs 63.1, within the ±7.8 margin).
- GPT-5.3-Codex (xhigh) is stronger in agents & tools, coding, instruction following, long context.
- GPT-5.4 Pro is stronger in multimodal, knowledge, reasoning.
- GPT-5.3-Codex (xhigh) is 14× cheaper ($4.81 vs $67.50 per 1M blended).
- GPT-5.3-Codex (xhigh) streams 16.3× faster (114 vs 7 tokens per second).
At a glance
| Spec | GPT-5.3-Codex | GPT-5.4 Pro |
|---|---|---|
| Developer | OpenAI | OpenAI |
| Released | 5 Feb 2026 | 4 Mar 2026 |
| Context window (tokens) | 400,000 | 1,050,000 |
| Compared at | GPT-5.3-Codex (xhigh) | GPT-5.4 Pro |
| Open weights | No | No |
| Price in / out / cached input per 1M | $1.75 / $14.00 / $0.175 | $30.00 / $180.00 / – |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | GPT-5.3-Codex (xhigh) | GPT-5.4 Pro | Leader |
|---|---|---|---|
| Reasoning | – | 72.7 | not comparable |
| Coding | 60.9 | – | not comparable |
| Agents & tools | 80.0 | – | not comparable |
| Knowledge | – | 73.6 | not comparable |
| Instruction following | 73.5 | 62.8 | GPT-5.3-Codex |
| Multimodal | 61.1 | 69.8 | GPT-5.4 Pro |
| Long context | 66.6 | – | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Dashed: that model has no result on the axis it crosses; the line only joins its neighbours.
Not drawn, no result for these models: Maths, Human preference.
Where sources disagree
Across the 0 benchmarks both have, GPT-5.3-Codex leads 0, GPT-5.4 Pro leads 0.
| Overall index | GPT-5.3-Codex (xhigh) | GPT-5.4 Pro | Leader |
|---|---|---|---|
| BenchLeader Index | 62.9 | 63.1 | neck and neck |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | GPT-5.3-Codex (xhigh) | GPT-5.4 Pro |
|---|---|---|
| Blended price per 1M tokens | $4.81 | $67.50 |
| Input / output per 1M | $1.75 / $14.00 | $30.00 / $180.00 |
| Cached input per 1M | $0.175 | – |
| Price per task: chat reply | $0.0049 | $0.066 |
| Price per task: summarize a 30-page report | $0.029 | $0.468 |
| Price per task: code edit | $0.032 | $0.450 |
| Price per task: agentic coding session | $0.161 | $2.52 |
| Price per task: structured extraction | $0.0063 | $0.096 |
| Agentic coding session, input mostly cached | $0.090 | – |
| Output speed | 114 tok/s | 7 tok/s |
| Time to first answer | 64.7 s | 8.2 s |
| Time to a complete answer | 69.1 s | – |
Each model is compared at its best-ranked reasoning-effort setting. GPT-5.3-Codex (xhigh) scores 62.9 and takes 65 s to its first answer token. GPT-5.4 Pro scores 63.1 and takes 8 s to its first answer token; GPT-5.4 Pro (xhigh) scores 59.9 in 8 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
41 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | GPT-5.3-Codex (xhigh) | GPT-5.4 Pro |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 62.9 | 63.1 |
| Agents & tools score | BenchLeader | 80.0 | – |
| Coding score | BenchLeader | 60.9 | – |
| Instruction following score | BenchLeader | 73.5 | 62.8 |
| Long context score | BenchLeader | 66.6 | – |
| Multimodal score | BenchLeader | 61.1 | 69.8 |
| Knowledge score | BenchLeader | – | 73.6 |
| Reasoning score | BenchLeader | – | 72.7 |
| Blended price $/M | BenchLeader | $4.81 | $67.50 |
| Output speed | BenchLeader | 114 tok/s | 7 tok/s |
| Time to first answer | BenchLeader | 64.7 s | 8.2 s |
| Context window | BenchLeader | 400,000 tokens | 1,050,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | – | 94.6%† |
| Humanity's Last Exam | Scale AI / CAIS | – | 44.3% |
| SimpleBench | SimpleBench | – | 74.1% |
| GPQA Diamond (AA) | Artificial Analysis | 91.5% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 42.5% | – |
| ARC-AGI-1 | ARC Prize | – | 94.5%† |
| ARC-AGI-2 | ARC Prize | – | 83.3%† |
| CritPt | Artificial Analysis | 16.9% | 30.0%† |
| Chess Puzzles | Epoch AI Benchmarking Hub | – | 59.0%† |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 74.8%† | – |
| WeirdML | WeirdML | 77.9% | 57.4%† |
| Arena WebDev | Arena (LMArena) | 1407† | – |
| LiveCodeBench | Vals AI | 87.3% | – |
| IOI | Vals AI | 53.8% | – |
| SWE-bench (Vals) | Vals AI | 78.0% | – |
| Vibe Code Bench v1.1 | Vals AI | 61.8% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1655.2 | – |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 78.4%† | – |
| Terminal-Bench Hard | Artificial Analysis | 53.0% | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 86.0% | – |
| HiL-Bench | Scale AI SEAL | 4.3%† | – |
| Terminal-Bench 2.0 (Vals) | Vals AI | 64.0%† | – |
| METR Time Horizons | METR | 74.5%† | – |
| Vending-Bench 2 | Andon Labs | 5940.1† | – |
| AA-Briefcase v1.1 | Artificial Analysis | 868 | – |
Show all 41 benchmarks
| Metric | Publisher | GPT-5.3-Codex (xhigh) | GPT-5.4 Pro |
|---|---|---|---|
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | – | 82.5%† |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | – | 58.5%† |
| MathArena Apex | MathArena | – | 69.8%† |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | – | 46.3%† |
| MultiNRC | Scale AI SEAL | – | 62.3% |
| Instruction following | |||
| IFBench | Artificial Analysis | 75.4% | – |
| MultiChallenge | Scale AI SEAL | – | 69.2% |
| TutorBench | Scale AI SEAL | – | 56.6% |
| Multimodal | |||
| MMMU-Pro | Artificial Analysis | 78.5% | – |
| VISTA | Scale AI SEAL | – | 53.9% |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 83.3% | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | – | 159.8 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 32.5 | – |
| AA Progress Index v1 | Artificial Analysis | 301.3 | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | – | 14.8% |
| MASK | Scale AI SEAL | – | 91.7% |
Data as of 2026-11-02. Best configuration of each model; every score links to its source on the model pages.
What changed
- GPT-5.3-Codex (xhigh): 1 new result (AA-Briefcase v1.1).
- GPT-5.3-Codex (xhigh): 1 new result (AA Progress Index v1).
- GPT-5.4 Pro: BenchLeader Index 63.6 → 63.1.
- GPT-5.4 Pro: 1 score revision.
- GPT-5.4 Pro: BenchLeader Index 62.1 → 63.6.
- GPT-5.4 Pro: 1 new result (MultiNRC).
- GPT-5.4 Pro: BenchLeader Index 64.0 → 62.1.
- GPT-5.4 Pro: BenchLeader Index 63.4 → 64.0.
- GPT-5.4 Pro: 1 new result (TutorBench).
- GPT-5.4 Pro: BenchLeader Index 64.0 → 63.4.
Related comparisons
- GPT-5.3-Codex vs Claude Opus 5.5
- GPT-5.3-Codex vs Gemini 4 Carbon
- GPT-5.3-Codex vs GPT-6 Astra
- GPT-5.4 Pro vs Claude Opus 5.5
- GPT-5.4 Pro vs Gemini 4 Carbon
- GPT-5.4 Pro vs GPT-6 Astra
- GPT-5.4 Pro vs GPT-6.1 Sol
- GPT-5.3-Codex vs Claude Fable 5.1
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
GPT-5.3-Codex vs GPT-5.4 Pro: questions
- Is GPT-5.3-Codex better than GPT-5.4 Pro?
- GPT-5.3-Codex (xhigh) and GPT-5.4 Pro are neck and neck on quality (62.9 vs 63.1, within the ±7.8 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 2 Nov 2026, so check the category scores for your use.
- GPT-5.3-Codex vs GPT-5.4 Pro: which is cheaper, and will that change?
- GPT-5.3-Codex is cheaper today: $4.81 against $67.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Which is faster, GPT-5.3-Codex or GPT-5.4 Pro?
- GPT-5.3-Codex streams 3×+ faster (114 vs 7 output tokens per second, Artificial Analysis 7-day median).
- Which is better for long documents?
- GPT-5.4 Pro accepts more context: 1,050,000 against 400,000 tokens for GPT-5.3-Codex.
- Which has lower latency?
- GPT-5.4 Pro answers first: 8.2 s against 64.7 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
More questions (1)
- When was GPT-5.4 Pro released?
- GPT-5.4 Pro was released on 4 Mar 2026.