Gemini 4 Neon vs GPT-6.1 Terra
Gemini 4 Neon leads GPT-6.1 Terra on coding, 86.9 to 76.6 on the Vibe Coding Index, and GPT-6.1 Terra costs 3.0x less per output token ($5.00 vs $15.00 per million). Updated November 22, 2026; 63 benchmarks have a result for both.
Gemini 4 Neon or GPT-6.1 Terra: which to pick
Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 76.6 for GPT-6.1 Terra.
Pick GPT-6.1 Terra for output tokens 3.0x cheaper ($5.00 vs $15.00 per million) and a larger context window (1.05M against 1.05M tokens).
- Shared results
- 63
- Gemini 4 Neon only
- 6
- GPT-6.1 Terra only
- 23
- Like-for-like categories
- 8 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
Gemini 4 Neon
Gemini 4 Neon scores 86.9 against 76.6.
4 benchmarks behind each score
Intelligence
AA Intelligence Index
Gemini 4 Neon
Gemini 4 Neon leads, 61.4 against 51.3.
Same board, both measured
Web design
Design Arena: websites
Gemini 4 Neon
Gemini 4 Neon leads, 1347 against 1283.
Same board, both measured
General
Vals Index
Gemini 4 Neon
Gemini 4 Neon leads, 70.9% against 60.5%.
Same board, both measured
Long documents
Listed context window
GPT-6.1 Terra
GPT-6.1 Terra takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6.1 Terra
Gemini 4 Neon $0.011, GPT-6.1 Terra $0.0035 per turn.
List prices
At a glance
- Coding index
- Gemini 4 Neon 86.9 from 4 benchmarks, GPT-6.1 Terra 76.6 from 4 benchmarks.
- Widest category
- Agentic work, 23.6 points apart on average over 8 shared benchmarks.
- Widest single gap
- Vending-Bench 2: Gemini 4 Neon $16,843 against GPT-6.1 Terra $9,107, 43.1 points apart on a 0 to 100 scale.
- Output price
- Gemini 4 Neon $15.00 against GPT-6.1 Terra $5.00 per million output tokens, 3.0x apart.
- Context window
- Gemini 4 Neon takes 1.05M tokens, GPT-6.1 Terra 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 | Gemini 4 Neon | Bars | GPT-6.1 Terra | Ties |
|---|---|---|---|---|
| Coding | 11 | 0 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 8 | 0 | ||
| Reasoning | 9 | 1 | ||
| Math | 4 | 0 | 1 tie | |
| Knowledge | 6 | 0 | ||
| Web design | 6 | 0 | ||
| General | 16 | 0 | ||
| All shared boards | 61 | 1 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index110
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| LiveBench coding | 89.7% | 89.2% | Gemini 4 Neon |
| LMArena WebDev | 1765 | 1685 | Gemini 4 Neon |
| Terminal-Bench 4 (Vals AI) | 61.6% | 49.5% | Gemini 4 Neon |
| Vibe Code Bench (Vals AI) | 93.7% | 84.0% | Gemini 4 Neon |
| SciCode | Not measured | 55.1% | Only GPT-6.1 Terra measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 63.6% | 55.6% | Gemini 4 Neon |
| FrontierCode 2.0 | 44.4% | 42.3% | Gemini 4 Neon |
| MirrorCode | Not measured | 76.7% | Only GPT-6.1 Terra measured |
| ALE-Bench | Not measured | 2608.8 | Only GPT-6.1 Terra measured |
| LMArena Text: coding | 1555 | 1482 | Gemini 4 Neon |
| LMArena Text: software and IT | 1542 | 1462 | Gemini 4 Neon |
| LMArena WebDev: React | 1787 | 1676 | Gemini 4 Neon |
| LMArena WebDev: HTML | 1803 | 1734 | Gemini 4 Neon |
| LMArena WebDev: from an image | Not measured | 1674 | Only GPT-6.1 Terra measured |
| GameBench 2 | 98.0 | 85.8 | Gemini 4 Neon |
| Terminal-Bench 4.0 | Not measured | 51.8% | Only GPT-6.1 Terra measured |
| CursorBench 4.0 | 65.4% | Not measured | Only Gemini 4 Neon measured |
| Neither model has a result on the other 35 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| AA Intelligence Index | 61.4 | 51.3 | Gemini 4 Neon |
| Epoch Capabilities Index | Not measured | 167.7 | Only GPT-6.1 Terra measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index80
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| Design Arena: agents building mobile apps | Not measured | 1166 | Only GPT-6.1 Terra measured |
| Design Arena: spreadsheets | Not measured | 992 | Only GPT-6.1 Terra measured |
| LiveBench research | 94.5% | 79.1% | Gemini 4 Neon |
| LiveBench browser | 95.6% | 93.1% | Gemini 4 Neon |
| LiveBench long agentic | 82.6% | 55.3% | Gemini 4 Neon |
| Tax Agent Bench (Vals AI) | 54.9% | 17.1% | Gemini 4 Neon |
| Legal Research Bench (Vals AI) | 56.3% | 26.0% | Gemini 4 Neon |
| Harvey's Legal Agent Benchmark (Vals AI) | 19.6% | 1.7% | Gemini 4 Neon |
| BioMysteryBench (Vals AI) | 80.0% | Not measured | Only Gemini 4 Neon measured |
| Terminal-Bench-Science (Vals AI) | 62.9% | Not measured | Only Gemini 4 Neon measured |
| Vals Science Index | 62.6% | Not measured | Only Gemini 4 Neon measured |
| APEX-Agents | 82.9% | 68.3% | Gemini 4 Neon |
| SpireClimb | Not measured | 22.3% | Only GPT-6.1 Terra measured |
| Vending-Bench 2 | $16,843 | $9,107 | Gemini 4 Neon |
| Neither model has a result on the other 30 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index91
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| LiveBench reasoning | 89.6% | 88.7% | Gemini 4 Neon |
| MysteryMechanism (Vals AI) | 45.0% | Not measured | Only Gemini 4 Neon measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 46.2% | Only GPT-6.1 Terra measured |
| HLE-Diamond (Artificial Analysis run) | 71.3% | 47.3% | Gemini 4 Neon |
| CritPt v2, hard subset (Artificial Analysis run) | 61.5% | 68.0% | GPT-6.1 Terra |
| GPQA Diamond (Epoch AI run) | 96.3% | 94.3% | Gemini 4 Neon |
| SimpleBench | 94.4% | 72.2% | Gemini 4 Neon |
| Chess puzzles | 83.0% | Not measured | Only Gemini 4 Neon measured |
| Mystery game puzzles | 85.0% | 74.0% | Gemini 4 Neon |
| EBR-bench | 98.1% | 78.1% | Gemini 4 Neon |
| LMArena Text: hard prompts | 1547 | 1451 | Gemini 4 Neon |
| LMArena Text: expert | 1560 | 1479 | Gemini 4 Neon |
| LMArena Text: hard prompts in English | 1524 | 1452 | Gemini 4 Neon |
| LMArena Vision: diagrams | Not measured | 1259 | Only GPT-6.1 Terra measured |
| LMArena Vision: homework | Not measured | 1308 | Only GPT-6.1 Terra measured |
| Humanity's Last Exam (Scale run) | Not measured | 35.7% | Only GPT-6.1 Terra measured |
| Neither model has a result on the other 10 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| OTIS Mock AIME | 100.0% | 100.0% | Tie |
| FrontierMath, tiers 1 to 3 | 91.2% | 90.9% | Gemini 4 Neon |
| FrontierMath, tier 4 | 97.6% | 92.7% | Gemini 4 Neon |
| LMArena Text: math | 1548 | 1450 | Gemini 4 Neon |
| LMArena Text: mathematical work | 1566 | 1477 | Gemini 4 Neon |
| Neither model has a result on the other 9 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| AA-Omniscience | 54.8 | 29.6 | Gemini 4 Neon |
| SimpleQA Verified | 75.2% | 57.3% | Gemini 4 Neon |
| LMArena Text: science | 1549 | 1422 | Gemini 4 Neon |
| LMArena Text: medicine and healthcare | 1534 | 1397 | Gemini 4 Neon |
| LMArena Text: legal and government | 1518 | 1455 | Gemini 4 Neon |
| LMArena Text: business and finance | 1493 | 1412 | Gemini 4 Neon |
| FACTS Benchmark Suite | Not measured | 74.1% | Only GPT-6.1 Terra measured |
| Neither model has a result on the other 14 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| Design Arena: websites | 1347 | 1283 | Gemini 4 Neon |
| Design Arena: UI components | 1349 | 1284 | Gemini 4 Neon |
| Design Arena: game dev | 1394 | 1316 | Gemini 4 Neon |
| Design Arena: data viz | Not measured | 1165 | Only GPT-6.1 Terra measured |
| Design Arena: 3D | 1490 | 1285 | Gemini 4 Neon |
| Design Arena: all code categories | 1391 | 1293 | Gemini 4 Neon |
| Design Arena: SVG | Not measured | 1298 | Only GPT-6.1 Terra measured |
| LiveBench design | 81.5% | 79.7% | Gemini 4 Neon |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index160
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | Leader |
|---|---|---|---|
| Vals Index | 70.9% | 60.5% | Gemini 4 Neon |
| LiveBench data analysis | 85.3% | 82.3% | Gemini 4 Neon |
| Long context reasoning (AA-LCR) | Not measured | 82.7% | Only GPT-6.1 Terra measured |
| GDPval-AA | Not measured | 1476 | Only GPT-6.1 Terra measured |
| Furniture assembly | 89.2% | 79.2% | Gemini 4 Neon |
| LMArena Text | 1524 | 1426 | Gemini 4 Neon |
| LMArena Text: instruction following | 1535 | 1442 | Gemini 4 Neon |
| LMArena Text: creative writing | 1519 | 1401 | Gemini 4 Neon |
| LMArena Vision | Not measured | 1247 | Only GPT-6.1 Terra measured |
| LMArena Text: English | 1511 | 1431 | Gemini 4 Neon |
| LMArena Text: non-English | 1524 | 1416 | Gemini 4 Neon |
| LMArena Text: Chinese | 1598 | 1466 | Gemini 4 Neon |
| LMArena Text: French | 1563 | 1445 | Gemini 4 Neon |
| LMArena Text: Spanish | 1505 | 1418 | Gemini 4 Neon |
| LMArena Text: Russian | 1554 | 1419 | Gemini 4 Neon |
| LMArena Text: multi-turn | 1525 | 1432 | Gemini 4 Neon |
| LMArena Text: longer queries | 1533 | 1445 | Gemini 4 Neon |
| LMArena Text: writing and language | 1517 | 1416 | Gemini 4 Neon |
| LMArena Text: entertainment and media | 1495 | 1414 | Gemini 4 Neon |
| LMArena Vision: English | Not measured | 1255 | Only GPT-6.1 Terra measured |
| LMArena Vision: reading text | Not measured | 1262 | Only GPT-6.1 Terra measured |
| LMArena Vision: creative writing | Not measured | 1215 | Only GPT-6.1 Terra measured |
| MultiChallenge | Not measured | 74.2% | Only GPT-6.1 Terra measured |
| Neither model has a result on the other 26 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemini 4 Neon | GPT-6.1 Terra | 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-6.1 Terra costs less
Repository review
50K input + 3K output tokens
GPT-6.1 Terra costs less
Agent loop
200K input + 10K output tokens
GPT-6.1 Terra 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 | Google Gemini 4 Neon | OpenAI GPT-6.1 Terra |
|---|---|---|
| Maker | OpenAI | |
| Weights | No public weights listed | No public weights listed |
| Released | November 18, 2026 | October 26, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $3.00 | $1.00 |
| Output $/Mtok | $15.00 | $5.00 |
| Coding index | 86.9 | 76.6 |
| Independent measurements behind it | 4 | 4 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemini 4 Neon better than GPT-6.1 Terra for coding?
Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 76.6 for GPT-6.1 Terra. Pick GPT-6.1 Terra for output tokens 3.0x cheaper ($5.00 vs $15.00 per million) and a larger context window (1.05M against 1.05M tokens).
Which is cheaper, Gemini 4 Neon or GPT-6.1 Terra?
Gemini 4 Neon costs $3.00 per million input tokens and $15.00 per million output tokens; GPT-6.1 Terra costs $1.00 and $5.00. Cost per solved task is known for neither of them.
Which has the bigger context window, Gemini 4 Neon or GPT-6.1 Terra?
GPT-6.1 Terra, with 1.05M tokens against 1.05M.
How many benchmarks is the Gemini 4 Neon vs GPT-6.1 Terra comparison based on?
63 benchmarks have a result for both models as of November 22, 2026. Gemini 4 Neon's coding index rests on 4 independent benchmarks and GPT-6.1 Terra's on 4. Where only one model has a result, the page says so instead of naming a winner.