GPT-5.2-Codex vs Muse Spark 1.3
Muse Spark 1.3 leads GPT-5.2-Codex on coding, 57.4 to 45.4 on the Vibe Coding Index, and Muse Spark 1.3 costs 3.3x less per output token ($4.25 vs $14.00 per million). Updated October 21, 2026; 13 benchmarks have a result for both.
GPT-5.2-Codex or Muse Spark 1.3: which to pick
Pick Muse Spark 1.3 for coding: it scores 57.4 on the Vibe Coding Index against 45.4 for GPT-5.2-Codex.
GPT-5.2-Codex is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
- Shared results
- 13
- GPT-5.2-Codex only
- 9
- Muse Spark 1.3 only
- 79
- Like-for-like categories
- 5 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
Muse Spark 1.3
Muse Spark 1.3 scores 57.4 against 45.4.
3 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Muse Spark 1.3
Muse Spark 1.3 leads, 48.1 against 28.5.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Muse Spark 1.3 has a AA Agentic Index result, 55.5.
Directional only
Web design
Design Arena: websites
No call
Only Muse Spark 1.3 has a Design Arena: websites result, 1356.
Directional only
Long documents
Listed context window
Muse Spark 1.3
Muse Spark 1.3 takes 1.05M tokens against 400K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Muse Spark 1.3
GPT-5.2-Codex $0.0088, Muse Spark 1.3 $0.0034 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
- GPT-5.2-Codex 45.4 from 3 benchmarks, Muse Spark 1.3 57.4 from 4 benchmarks.
- Widest category
- Coding, 38.1 points apart on average over 5 shared benchmarks.
- Widest single gap
- LMArena WebDev: GPT-5.2-Codex 1339 against Muse Spark 1.3 1659, 45.8 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2-Codex $14.00 against Muse Spark 1.3 $4.25 per million output tokens, 3.3x apart.
- Context window
- GPT-5.2-Codex takes 400K tokens, Muse Spark 1.3 1.05M.
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 | GPT-5.2-Codex | Bars | Muse Spark 1.3 | Ties |
|---|---|---|---|---|
| Coding | 0 | 5 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 4 | ||
| Reasoning | 0 | 2 | ||
| General | 0 | 1 | ||
| All shared boards | 0 | 13 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index05
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| SWE-bench Verified | 72.8% | Not measured | Only GPT-5.2-Codex measured |
| SWE-bench Multilingual | 66.3% | Not measured | Only GPT-5.2-Codex measured |
| LiveBench coding | Not measured | 90.2% | Only Muse Spark 1.3 measured |
| LMArena WebDev | 1339 | 1659 | Muse Spark 1.3 |
| AA Coding Index | Not measured | 75.8 | Only Muse Spark 1.3 measured |
| LiveCodeBench (Vals AI) | 88.0% | Not measured | Only GPT-5.2-Codex measured |
| SWE-bench Verified (Vals AI) | 72.4% | Not measured | Only GPT-5.2-Codex measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 10.6% | Only Muse Spark 1.3 measured |
| Vibe Code Bench (Vals AI) | 37.9% | 82.9% | Muse Spark 1.3 |
| Code migration (Vals AI) | Not measured | 27.6% | Only Muse Spark 1.3 measured |
| ProgramBench (Vals AI) | Not measured | 0.5% | Only Muse Spark 1.3 measured |
| IOI (Vals AI) | Not measured | 43.9% | Only Muse Spark 1.3 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 72.3% | Only Muse Spark 1.3 measured |
| CyberBench (Vals AI) | Not measured | 69.4% | Only Muse Spark 1.3 measured |
| SciCode | Not measured | 59.7% | Only Muse Spark 1.3 measured |
| Terminal-Bench Hard (Artificial Analysis run) | 37.1% | Not measured | Only GPT-5.2-Codex measured |
| Terminal-Bench | 66.5% | Not measured | Only GPT-5.2-Codex measured |
| FrontierSWE | Not measured | 25.8% | Only Muse Spark 1.3 measured |
| ALE-Bench | 1299.9 | 1688.3 | Muse Spark 1.3 |
| LMArena Text: coding | Not measured | 1510 | Only Muse Spark 1.3 measured |
| LMArena Text: software and IT | Not measured | 1511 | Only Muse Spark 1.3 measured |
| LMArena WebDev: React | 1329 | 1672 | Muse Spark 1.3 |
| LMArena WebDev: HTML | 1348 | 1653 | Muse Spark 1.3 |
| LMArena WebDev: from an image | Not measured | 1640 | Only Muse Spark 1.3 measured |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 28.5 | 48.1 | Muse Spark 1.3 |
| Epoch Capabilities Index | Not measured | 157.2 | Only Muse Spark 1.3 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 55.5 | Only Muse Spark 1.3 measured |
| Design Arena: agents building web apps | 1051 | 1297 | Muse Spark 1.3 |
| Design Arena: agents building full-stack apps | 993 | 1295 | Muse Spark 1.3 |
| Design Arena: agents building mobile apps | 1090 | 1217 | Muse Spark 1.3 |
| Design Arena: agents building Android apps | 1176 | Not measured | Only GPT-5.2-Codex measured |
| Design Arena: agents building Godot games | 1142 | 1241 | Muse Spark 1.3 |
| Design Arena: PowerPoint slides | Not measured | 1243 | Only Muse Spark 1.3 measured |
| Design Arena: spreadsheets | Not measured | 1043 | Only Muse Spark 1.3 measured |
| LiveBench research | Not measured | 68.6% | Only Muse Spark 1.3 measured |
| LiveBench browser | Not measured | 80.0% | Only Muse Spark 1.3 measured |
| LiveBench long agentic | Not measured | 40.4% | Only Muse Spark 1.3 measured |
| Finance Agent (Vals AI) | Not measured | 58.9% | Only Muse Spark 1.3 measured |
| Tax Agent Bench (Vals AI) | Not measured | 36.0% | Only Muse Spark 1.3 measured |
| Legal Research Bench (Vals AI) | Not measured | 40.9% | Only Muse Spark 1.3 measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 62.7% | Only Muse Spark 1.3 measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 4.3% | Only Muse Spark 1.3 measured |
| τ²-Bench Telecom | 92.1% | Not measured | Only GPT-5.2-Codex measured |
| APEX-Agents | Not measured | 58.6% | Only Muse Spark 1.3 measured |
| Neither model has a result on the other 23 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| LiveBench reasoning | Not measured | 89.7% | Only Muse Spark 1.3 measured |
| GPQA Diamond (Artificial Analysis run) | 89.9% | 94.1% | Muse Spark 1.3 |
| Humanity's Last Exam (Artificial Analysis run) | 35.7% | 48.7% | Muse Spark 1.3 |
| Chess puzzles | Not measured | 37.0% | Only Muse Spark 1.3 measured |
| Mystery game puzzles | Not measured | 25.0% | Only Muse Spark 1.3 measured |
| LMArena Text: hard prompts | Not measured | 1503 | Only Muse Spark 1.3 measured |
| LMArena Text: expert | Not measured | 1519 | Only Muse Spark 1.3 measured |
| LMArena Text: hard prompts in English | Not measured | 1500 | Only Muse Spark 1.3 measured |
| LMArena Vision: diagrams | Not measured | 1330 | Only Muse Spark 1.3 measured |
| LMArena Vision: homework | Not measured | 1296 | Only Muse Spark 1.3 measured |
| Neither model has a result on the other 12 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| OTIS Mock AIME | Not measured | 99.2% | Only Muse Spark 1.3 measured |
| FrontierMath, tiers 1 to 3 | Not measured | 74.4% | Only Muse Spark 1.3 measured |
| FrontierMath, tier 4 | Not measured | 46.3% | Only Muse Spark 1.3 measured |
| ProofBench | Not measured | 58.0% | Only Muse Spark 1.3 measured |
| LMArena Text: math | Not measured | 1497 | Only Muse Spark 1.3 measured |
| LMArena Text: mathematical work | Not measured | 1491 | Only Muse Spark 1.3 measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| LMArena Text: science | Not measured | 1507 | Only Muse Spark 1.3 measured |
| LMArena Text: medicine and healthcare | Not measured | 1493 | Only Muse Spark 1.3 measured |
| LMArena Text: legal and government | Not measured | 1499 | Only Muse Spark 1.3 measured |
| LMArena Text: business and finance | Not measured | 1490 | Only Muse Spark 1.3 measured |
| Neither model has a result on the other 14 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1356 | Only Muse Spark 1.3 measured |
| Design Arena: UI components | Not measured | 1351 | Only Muse Spark 1.3 measured |
| Design Arena: game dev | Not measured | 1329 | Only Muse Spark 1.3 measured |
| Design Arena: data viz | Not measured | 1343 | Only Muse Spark 1.3 measured |
| Design Arena: 3D | Not measured | 1359 | Only Muse Spark 1.3 measured |
| Design Arena: all code categories | Not measured | 1353 | Only Muse Spark 1.3 measured |
| Design Arena: SVG | Not measured | 1301 | Only Muse Spark 1.3 measured |
| Design Arena: ASCII art | Not measured | 1280 | Only Muse Spark 1.3 measured |
| LiveBench design | Not measured | 74.8% | Only Muse Spark 1.3 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 79.6% | Only Muse Spark 1.3 measured |
| IFBench | 77.6% | Not measured | Only GPT-5.2-Codex measured |
| Long context reasoning (AA-LCR) | 82.3% | 83.0% | Muse Spark 1.3 |
| DTBench | Not measured | 96.5% | Only Muse Spark 1.3 measured |
| LMCA | Not measured | 53.9% | Only Muse Spark 1.3 measured |
| LMArena Text | Not measured | 1490 | Only Muse Spark 1.3 measured |
| LMArena Text: instruction following | Not measured | 1479 | Only Muse Spark 1.3 measured |
| LMArena Text: creative writing | Not measured | 1462 | Only Muse Spark 1.3 measured |
| LMArena Vision | Not measured | 1308 | Only Muse Spark 1.3 measured |
| LMArena Text: English | Not measured | 1495 | Only Muse Spark 1.3 measured |
| LMArena Text: non-English | Not measured | 1481 | Only Muse Spark 1.3 measured |
| LMArena Text: Chinese | Not measured | 1532 | Only Muse Spark 1.3 measured |
| LMArena Text: German | Not measured | 1513 | Only Muse Spark 1.3 measured |
| LMArena Text: Japanese | Not measured | 1470 | Only Muse Spark 1.3 measured |
| LMArena Text: Korean | Not measured | 1490 | Only Muse Spark 1.3 measured |
| LMArena Text: French | Not measured | 1511 | Only Muse Spark 1.3 measured |
| LMArena Text: Spanish | Not measured | 1496 | Only Muse Spark 1.3 measured |
| LMArena Text: Russian | Not measured | 1491 | Only Muse Spark 1.3 measured |
| LMArena Text: Polish | Not measured | 1490 | Only Muse Spark 1.3 measured |
| LMArena Text: multi-turn | Not measured | 1487 | Only Muse Spark 1.3 measured |
| LMArena Text: longer queries | Not measured | 1488 | Only Muse Spark 1.3 measured |
| LMArena Text: writing and language | Not measured | 1470 | Only Muse Spark 1.3 measured |
| LMArena Text: entertainment and media | Not measured | 1457 | Only Muse Spark 1.3 measured |
| LMArena Vision: English | Not measured | 1298 | Only Muse Spark 1.3 measured |
| LMArena Vision: Chinese | Not measured | 1384 | Only Muse Spark 1.3 measured |
| LMArena Vision: reading text | Not measured | 1319 | Only Muse Spark 1.3 measured |
| LMArena Vision: humor | Not measured | 1293 | Only Muse Spark 1.3 measured |
| LMArena Vision: creative writing | Not measured | 1296 | Only Muse Spark 1.3 measured |
| Neither model has a result on the other 18 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.2-Codex | Muse Spark 1.3 | 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
Muse Spark 1.3 costs less
Repository review
50K input + 3K output tokens
Muse Spark 1.3 costs less
Agent loop
200K input + 10K output tokens
Muse Spark 1.3 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 | OpenAI GPT-5.2-Codex | Meta Muse Spark 1.3 |
|---|---|---|
| Maker | OpenAI | Meta |
| Weights | No public weights listed | No public weights listed |
| Released | January 14, 2026 | September 2, 2026 |
| Context window | 400K | 1.05M |
| Input $/Mtok | $1.75 | $1.25 |
| Output $/Mtok | $14.00 | $4.25 |
| Coding index | 45.4 | 57.4 |
| Independent measurements behind it | 3 | 4 |
| Cost per solved task | $0.43 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2-Codex better than Muse Spark 1.3 for coding?
Pick Muse Spark 1.3 for coding: it scores 57.4 on the Vibe Coding Index against 45.4 for GPT-5.2-Codex. GPT-5.2-Codex is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
Which is cheaper, GPT-5.2-Codex or Muse Spark 1.3?
GPT-5.2-Codex costs $1.75 per million input tokens and $14.00 per million output tokens; Muse Spark 1.3 costs $1.25 and $4.25. Cost per solved task is known for GPT-5.2-Codex only.
Which has the bigger context window, GPT-5.2-Codex or Muse Spark 1.3?
Muse Spark 1.3, with 1.05M tokens against 400K.
How many benchmarks is the GPT-5.2-Codex vs Muse Spark 1.3 comparison based on?
13 benchmarks have a result for both models as of October 21, 2026. GPT-5.2-Codex's coding index rests on 3 independent benchmarks and Muse Spark 1.3's on 4. Where only one model has a result, the page says so instead of naming a winner.