Claude Opus 4.1 vs GPT-5.2-Codex
GPT-5.2-Codex leads Claude Opus 4.1 on the Artificial Analysis Intelligence Index, 28.5 to 22.8, and GPT-5.2-Codex costs 5.4x less per output token ($14.00 vs $75.00 per million). Updated October 22, 2026; 5 benchmarks have a result for both.
Claude Opus 4.1 or GPT-5.2-Codex: which to pick
Only GPT-5.2-Codex has a coding score, 45.4 on the Vibe Coding Index; Claude Opus 4.1 has none, so the two cannot be ranked on coding.
Pick Claude Opus 4.1 for work measured by LMArena WebDev or LMArena WebDev: HTML, where it leads.
- Shared results
- 5
- Claude Opus 4.1 only
- 68
- GPT-5.2-Codex only
- 17
- Like-for-like categories
- 2 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Only GPT-5.2-Codex has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
GPT-5.2-Codex
GPT-5.2-Codex leads, 28.5 against 22.8.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Claude Opus 4.1 has a GPQA Diamond result, 76.3%.
Directional only
Math
AIME
No call
Only Claude Opus 4.1 has a AIME result, 78.2%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Opus 4.1 has a MMLU Pro result, 87.9%.
Directional only
Web design
Design Arena: websites
No call
Only Claude Opus 4.1 has a Design Arena: websites result, 1176.
Directional only
Long documents
Listed context window
GPT-5.2-Codex
GPT-5.2-Codex takes 400K tokens against 200K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-5.2-Codex
Claude Opus 4.1 $0.052, GPT-5.2-Codex $0.0088 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GPT-5.2-Codex has the number, $0.43.
Directional only
At a glance
- Coding index
- Claude Opus 4.1 not benchmarked, GPT-5.2-Codex 45.4 from 3 benchmarks.
- Widest category
- Coding, 16.4 points apart on average over 4 shared benchmarks.
- Widest single gap
- Terminal-Bench: Claude Opus 4.1 38.0% against GPT-5.2-Codex 66.5%, 28.5 points apart on a 0 to 100 scale.
- Output price
- Claude Opus 4.1 $75.00 against GPT-5.2-Codex $14.00 per million output tokens, 5.4x apart.
- Context window
- Claude Opus 4.1 takes 200K tokens, GPT-5.2-Codex 400K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Claude Opus 4.1 | Bars | GPT-5.2-Codex | Ties |
|---|---|---|---|---|
| Coding | 2 | 2 | ||
| Intelligence | 0 | 1 | ||
| All shared boards | 2 | 3 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index22
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| SWE-bench Verified | Not measured | 72.8% | Only GPT-5.2-Codex measured |
| SWE-bench Multilingual | Not measured | 66.3% | Only GPT-5.2-Codex measured |
| LMArena WebDev | 1390 | 1339 | Claude Opus 4.1 |
| LiveCodeBench (Vals AI) | 66.5% | 88.0% | GPT-5.2-Codex |
| SWE-bench Verified (Vals AI) | Not measured | 72.4% | Only GPT-5.2-Codex measured |
| Vibe Code Bench (Vals AI) | Not measured | 37.9% | Only GPT-5.2-Codex measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 37.1% | Only GPT-5.2-Codex measured |
| Terminal-Bench | 38.0% | 66.5% | GPT-5.2-Codex |
| WeirdML | 45.9% | Not measured | Only Claude Opus 4.1 measured |
| Cybench | 42.0% | Not measured | Only Claude Opus 4.1 measured |
| ALE-Bench | Not measured | 1299.9 | Only GPT-5.2-Codex measured |
| LMArena Text: coding | 1479 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: software and IT | 1456 | Not measured | Only Claude Opus 4.1 measured |
| LMArena WebDev: React | Not measured | 1329 | Only GPT-5.2-Codex measured |
| LMArena WebDev: HTML | 1399 | 1348 | Claude Opus 4.1 |
| SciCode | 4.6% | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 23 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| AA Intelligence Index | 22.8 | 28.5 | GPT-5.2-Codex |
| Epoch Capabilities Index | 144.1 | Not measured | Only Claude Opus 4.1 measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | Not measured | 1051 | Only GPT-5.2-Codex measured |
| Design Arena: agents building full-stack apps | Not measured | 993 | Only GPT-5.2-Codex measured |
| Design Arena: agents building mobile apps | Not measured | 1090 | Only GPT-5.2-Codex measured |
| Design Arena: agents building Android apps | Not measured | 1176 | Only GPT-5.2-Codex measured |
| Design Arena: agents building Godot games | Not measured | 1142 | Only GPT-5.2-Codex measured |
| τ²-Bench Telecom | Not measured | 92.1% | Only GPT-5.2-Codex measured |
| METR time horizons | 66.8% | Not measured | Only Claude Opus 4.1 measured |
| GDPval | 43.6% | Not measured | Only Claude Opus 4.1 measured |
| DeepResearch Bench | 48.3% | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 32 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| GPQA Diamond | 76.3% | Not measured | Only Claude Opus 4.1 measured |
| MMMU Pro (Vals AI) | 77.5% | Not measured | Only Claude Opus 4.1 measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 89.9% | Only GPT-5.2-Codex measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 35.7% | Only GPT-5.2-Codex measured |
| GPQA Diamond (Epoch AI run) | 77.3% | Not measured | Only Claude Opus 4.1 measured |
| SimpleBench | 60.0% | Not measured | Only Claude Opus 4.1 measured |
| Mystery game puzzles | 21.0% | Not measured | Only Claude Opus 4.1 measured |
| EBR-bench | 7.9% | Not measured | Only Claude Opus 4.1 measured |
| VPCT | 35.0% | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: hard prompts | 1442 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: expert | 1438 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: hard prompts in English | 1451 | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 10 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| AIME | 78.2% | Not measured | Only Claude Opus 4.1 measured |
| MATH 500 (Vals AI) | 95.4% | Not measured | Only Claude Opus 4.1 measured |
| MGSM (Vals AI) | 94.4% | Not measured | Only Claude Opus 4.1 measured |
| OTIS Mock AIME | 68.9% | Not measured | Only Claude Opus 4.1 measured |
| FrontierMath, tiers 1 to 3 | 12.6% | Not measured | Only Claude Opus 4.1 measured |
| FrontierMath, tier 4 | 2.4% | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: math | 1431 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: mathematical work | 1434 | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 5 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index11 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| MMLU Pro | 87.9% | Not measured | Only Claude Opus 4.1 measured |
| LegalBench (Vals AI) | 83.5% | Not measured | Only Claude Opus 4.1 measured |
| TaxEval (Vals AI) | 73.7% | Not measured | Only Claude Opus 4.1 measured |
| MortgageTax (Vals AI) | 61.1% | Not measured | Only Claude Opus 4.1 measured |
| MedQA (Vals AI) | 93.6% | Not measured | Only Claude Opus 4.1 measured |
| MedCode (Vals AI) | 47.2% | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: science | 1422 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: medicine and healthcare | 1416 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: legal and government | 1421 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: business and finance | 1406 | Not measured | Only Claude Opus 4.1 measured |
| FACTS Benchmark Suite | 40.4% | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 7 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| Design Arena: websites | 1176 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: UI components | 1162 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: game dev | 1171 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: data viz | 1166 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: 3D | 1152 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: all code categories | 1170 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: SVG | 1152 | Not measured | Only Claude Opus 4.1 measured |
| Design Arena: ASCII art | 1167 | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 1 board here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index24 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 31.4% | Not measured | Only Claude Opus 4.1 measured |
| MedScribe (Vals AI) | 73.9% | Not measured | Only Claude Opus 4.1 measured |
| IFBench | Not measured | 77.6% | Only GPT-5.2-Codex measured |
| Long context reasoning (AA-LCR) | Not measured | 82.3% | Only GPT-5.2-Codex measured |
| DTBench | 80.0% | Not measured | Only Claude Opus 4.1 measured |
| LMCA | 37.1% | Not measured | Only Claude Opus 4.1 measured |
| Creative writing (Lech Mazur) | 84.7% | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text | 1419 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: instruction following | 1435 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: creative writing | 1412 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: English | 1433 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: non-English | 1405 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: Chinese | 1427 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: German | 1413 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: Japanese | 1377 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: Korean | 1381 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: French | 1433 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: Spanish | 1448 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: Russian | 1422 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: Polish | 1413 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: multi-turn | 1444 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: longer queries | 1455 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: writing and language | 1418 | Not measured | Only Claude Opus 4.1 measured |
| LMArena Text: entertainment and media | 1398 | Not measured | Only Claude Opus 4.1 measured |
| Neither model has a result on the other 22 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Claude Opus 4.1 | GPT-5.2-Codex | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GPT-5.2-Codex costs less
Repository review
50K input + 3K output tokens
GPT-5.2-Codex costs less
Agent loop
200K input + 10K output tokens
GPT-5.2-Codex costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Anthropic Claude Opus 4.1 | OpenAI GPT-5.2-Codex |
|---|---|---|
| Maker | Anthropic | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | August 5, 2025 | January 14, 2026 |
| Context window | 200K | 400K |
| Input $/Mtok | $15.00 | $1.75 |
| Output $/Mtok | $75.00 | $14.00 |
| Coding index | Not benchmarked | 45.4 |
| Independent measurements behind it | 1 | 3 |
| Cost per solved task | Not computable | $0.43 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Claude Opus 4.1 better than GPT-5.2-Codex for coding?
Only GPT-5.2-Codex has a coding score, 45.4 on the Vibe Coding Index; Claude Opus 4.1 has none, so the two cannot be ranked on coding. Pick Claude Opus 4.1 for work measured by LMArena WebDev or LMArena WebDev: HTML, where it leads.
Which is cheaper, Claude Opus 4.1 or GPT-5.2-Codex?
Claude Opus 4.1 costs $15.00 per million input tokens and $75.00 per million output tokens; GPT-5.2-Codex costs $1.75 and $14.00. Cost per solved task is known for GPT-5.2-Codex only.
Which has the bigger context window, Claude Opus 4.1 or GPT-5.2-Codex?
GPT-5.2-Codex, with 400K tokens against 200K.
How many benchmarks is the Claude Opus 4.1 vs GPT-5.2-Codex comparison based on?
5 benchmarks have a result for both models as of October 22, 2026. Claude Opus 4.1's coding index rests on 1 independent benchmark and GPT-5.2-Codex's on 3. Where only one model has a result, the page says so instead of naming a winner.