GPT-6 Luna vs Mistral Large 4
GPT-6 Luna leads Mistral Large 4 on coding, 57.3 to 56.4 on the Vibe Coding Index, and GPT-6 Luna costs 8.4x less per output token ($0.50 vs $4.18 per million). Updated October 22, 2026; 79 benchmarks have a result for both.
GPT-6 Luna or Mistral Large 4: which to pick
GPT-6 Luna and Mistral Large 4 are level for coding, 57.3 against 56.4 on the Vibe Coding Index, so price and context decide.
Pick Mistral Large 4 for work measured by Terminal-Bench 4 (Vals AI) or LMArena Text: coding, where it leads. Pick GPT-6 Luna for output tokens 8.4x cheaper ($0.50 vs $4.18 per million) and a larger context window (1.05M against 1.05M tokens).
- Shared results
- 79
- GPT-6 Luna only
- 24
- Mistral Large 4 only
- 2
- 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
No call
Level: 57.3 against 56.4, inside 2 points.
5 benchmarks behind each score
Intelligence
AA Intelligence Index
Mistral Large 4
Mistral Large 4 leads, 38.4 against 38.1.
Same board, both measured
Web design
Design Arena: websites
GPT-6 Luna
GPT-6 Luna leads, 1271 against 1257.
Same board, both measured
General
Vals Index
No call
Only GPT-6 Luna has a Vals Index result, 52.3%.
Directional only
Long documents
Listed context window
GPT-6 Luna
GPT-6 Luna takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6 Luna
GPT-6 Luna $0.0003, Mistral Large 4 $0.0034 per turn.
List prices
At a glance
- Coding index
- GPT-6 Luna 57.3 from 5 benchmarks, Mistral Large 4 56.4 from 5 benchmarks.
- Widest category
- Math, 20.0 points apart on average over 6 shared benchmarks.
- Widest single gap
- LiveBench browser: GPT-6 Luna 72.0% against Mistral Large 4 8.0%, 64.0 points apart on a 0 to 100 scale.
- Output price
- GPT-6 Luna $0.50 against Mistral Large 4 $4.18 per million output tokens, 8.4x apart.
- Context window
- GPT-6 Luna takes 1.05M tokens, Mistral Large 4 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-6 Luna | Bars | Mistral Large 4 | Ties |
|---|---|---|---|---|
| Coding | 11 | 3 | ||
| Intelligence | 1 | 1 | ||
| Agentic work | 4 | 5 | ||
| Reasoning | 4 | 6 | ||
| Math | 4 | 2 | ||
| Knowledge | 2 | 4 | ||
| Web design | 6 | 3 | ||
| General | 5 | 18 | ||
| All shared boards | 37 | 42 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index113
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| LiveBench coding | 84.5% | 60.0% | GPT-6 Luna |
| LMArena WebDev | 1578 | 1543 | GPT-6 Luna |
| Terminal-Bench 4 (Vals AI) | 13.6% | 27.8% | Mistral Large 4 |
| Vibe Code Bench (Vals AI) | 81.7% | 78.4% | GPT-6 Luna |
| Code migration (Vals AI) | 42.5% | 30.6% | GPT-6 Luna |
| ProgramBench (Vals AI) | 0.5% | Not measured | Only GPT-6 Luna measured |
| IOI (Vals AI) | 55.6% | 45.3% | GPT-6 Luna |
| Terminal-Bench 2.1 (Vals AI) | 73.0% | Not measured | Only GPT-6 Luna measured |
| SWE-bench Multimodal (Vals AI) | 37.5% | 33.1% | GPT-6 Luna |
| Vibe Code Bench 1-100 (Vals AI) | Not measured | 14.3% | Only Mistral Large 4 measured |
| CyberBench (Vals AI) | 76.3% | Not measured | Only GPT-6 Luna measured |
| SciCode | 54.6% | 54.2% | GPT-6 Luna |
| FrontierCode | 42.4% | 39.4% | GPT-6 Luna |
| ALE-Bench | 1576.9 | 1213.2 | GPT-6 Luna |
| LMArena Text: coding | 1437 | 1478 | Mistral Large 4 |
| LMArena Text: software and IT | 1423 | 1467 | Mistral Large 4 |
| LMArena WebDev: React | 1573 | 1559 | GPT-6 Luna |
| LMArena WebDev: HTML | 1595 | 1487 | GPT-6 Luna |
| LMArena WebDev: from an image | 1571 | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| AA Intelligence Index | 38.1 | 38.4 | Mistral Large 4 |
| Epoch Capabilities Index | 156.8 | 150.0 | GPT-6 Luna |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index45
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1191 | Not measured | Only GPT-6 Luna measured |
| Design Arena: agents building full-stack apps | 1158 | Not measured | Only GPT-6 Luna measured |
| Design Arena: agents building mobile apps | 1134 | Not measured | Only GPT-6 Luna measured |
| LiveBench research | 43.5% | 28.1% | GPT-6 Luna |
| LiveBench browser | 72.0% | 8.0% | GPT-6 Luna |
| LiveBench long agentic | 48.1% | 11.8% | GPT-6 Luna |
| SREBench (Vals AI) | 2.7% | Not measured | Only GPT-6 Luna measured |
| Finance Agent (Vals AI) | 49.9% | 54.7% | Mistral Large 4 |
| Tax Agent Bench (Vals AI) | 19.9% | 30.4% | Mistral Large 4 |
| Legal Research Bench (Vals AI) | 30.3% | 31.7% | Mistral Large 4 |
| Harvey's Legal Agent Benchmark (Vals AI) | 4.2% | 17.5% | Mistral Large 4 |
| Excel Modeling Benchmark (Vals AI) | 68.5% | 56.2% | GPT-6 Luna |
| BioMysteryBench (Vals AI) | 61.5% | 67.0% | Mistral Large 4 |
| Terminal-Bench-Science (Vals AI) | 4.3% | Not measured | Only GPT-6 Luna measured |
| Vals Science Index | 28.4% | Not measured | Only GPT-6 Luna measured |
| APEX-Agents | 44.3% | Not measured | Only GPT-6 Luna measured |
| SpireClimb | 0.0% | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index46
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| LiveBench reasoning | 81.8% | 83.9% | Mistral Large 4 |
| MysteryMechanism (Vals AI) | 19.4% | 12.2% | GPT-6 Luna |
| Humanity's Last Exam (Artificial Analysis run) | 38.5% | 35.0% | GPT-6 Luna |
| GPQA Diamond (Epoch AI run) | 90.5% | 88.1% | GPT-6 Luna |
| ARC-AGI | 86.7% | Not measured | Only GPT-6 Luna measured |
| ARC-AGI-2 | 59.3% | Not measured | Only GPT-6 Luna measured |
| Chess puzzles | 36.0% | 33.0% | GPT-6 Luna |
| Mystery game puzzles | 7.0% | 17.0% | Mistral Large 4 |
| LMArena Text: hard prompts | 1410 | 1444 | Mistral Large 4 |
| LMArena Text: expert | 1442 | 1444 | Mistral Large 4 |
| LMArena Text: hard prompts in English | 1410 | 1442 | Mistral Large 4 |
| LMArena Vision: diagrams | 1245 | 1288 | Mistral Large 4 |
| LMArena Vision: homework | 1261 | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 9 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index42
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.9% | 98.3% | GPT-6 Luna |
| FrontierMath, tiers 1 to 3 | 78.9% | 58.6% | GPT-6 Luna |
| FrontierMath, tier 4 | 56.1% | 26.8% | GPT-6 Luna |
| ProofBench | 64.0% | 10.0% | GPT-6 Luna |
| LMArena Text: math | 1409 | 1469 | Mistral Large 4 |
| LMArena Text: mathematical work | 1427 | 1462 | Mistral Large 4 |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index24
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| MedCode (Vals AI) | 44.7% | 40.7% | GPT-6 Luna |
| SimpleQA Verified | 41.4% | 20.0% | GPT-6 Luna |
| LMArena Text: science | 1386 | 1448 | Mistral Large 4 |
| LMArena Text: medicine and healthcare | 1352 | 1433 | Mistral Large 4 |
| LMArena Text: legal and government | 1400 | 1425 | Mistral Large 4 |
| LMArena Text: business and finance | 1388 | 1439 | Mistral Large 4 |
| FACTS Benchmark Suite | 63.3% | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 11 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index63
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| Design Arena: websites | 1271 | 1257 | GPT-6 Luna |
| Design Arena: UI components | 1295 | 1272 | GPT-6 Luna |
| Design Arena: game dev | 1257 | 1225 | GPT-6 Luna |
| Design Arena: data viz | 1216 | 1275 | Mistral Large 4 |
| Design Arena: 3D | 1218 | 1252 | Mistral Large 4 |
| Design Arena: all code categories | 1270 | 1259 | GPT-6 Luna |
| Design Arena: SVG | 1252 | 1232 | GPT-6 Luna |
| Design Arena: ASCII art | 1153 | 1202 | Mistral Large 4 |
| LiveBench design | 82.5% | 65.6% | GPT-6 Luna |
GeneralLike-for-likePublic boards, shown as published, kept out of the index518
| Benchmark | GPT-6 Luna | Mistral Large 4 | Leader |
|---|---|---|---|
| Vals Index | 52.3% | Not measured | Only GPT-6 Luna measured |
| LiveBench data analysis | 73.4% | 76.5% | Mistral Large 4 |
| Public Benefits Bench (Vals AI) | 57.6% | Not measured | Only GPT-6 Luna measured |
| SAGE (Vals AI) | 48.1% | Not measured | Only GPT-6 Luna measured |
| MedScribe (Vals AI) | 83.7% | 80.4% | GPT-6 Luna |
| Long context reasoning (AA-LCR) | 83.3% | 81.3% | GPT-6 Luna |
| DTBench | 90.1% | 88.2% | GPT-6 Luna |
| LMCA | 44.5% | 35.8% | GPT-6 Luna |
| Surface Evolver Bench | Not measured | 28.1% | Only Mistral Large 4 measured |
| Furniture assembly | 44.2% | 20.8% | GPT-6 Luna |
| LMArena Text | 1389 | 1431 | Mistral Large 4 |
| LMArena Text: instruction following | 1408 | 1422 | Mistral Large 4 |
| LMArena Text: creative writing | 1363 | 1363 | Mistral Large 4 |
| LMArena Vision | 1216 | 1272 | Mistral Large 4 |
| LMArena Text: English | 1393 | 1433 | Mistral Large 4 |
| LMArena Text: non-English | 1381 | 1422 | Mistral Large 4 |
| LMArena Text: Chinese | 1427 | 1492 | Mistral Large 4 |
| LMArena Text: German | 1376 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: Japanese | 1369 | 1415 | Mistral Large 4 |
| LMArena Text: Korean | 1359 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: French | 1408 | 1492 | Mistral Large 4 |
| LMArena Text: Spanish | 1391 | 1455 | Mistral Large 4 |
| LMArena Text: Russian | 1392 | 1417 | Mistral Large 4 |
| LMArena Text: Polish | 1383 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: multi-turn | 1392 | 1435 | Mistral Large 4 |
| LMArena Text: longer queries | 1408 | 1428 | Mistral Large 4 |
| LMArena Text: writing and language | 1384 | 1386 | Mistral Large 4 |
| LMArena Text: entertainment and media | 1364 | 1378 | Mistral Large 4 |
| LMArena Vision: English | 1204 | 1283 | Mistral Large 4 |
| LMArena Vision: Chinese | 1266 | Not measured | Only GPT-6 Luna measured |
| LMArena Vision: reading text | 1225 | 1284 | Mistral Large 4 |
| LMArena Vision: creative writing | 1205 | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 14 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-6 Luna | Mistral Large 4 | 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 Luna costs less
Repository review
50K input + 3K output tokens
GPT-6 Luna costs less
Agent loop
200K input + 10K output tokens
GPT-6 Luna 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-6 Luna | Mistral Large 4 |
|---|---|---|
| Maker | OpenAI | Mistral AI |
| Weights | No public weights listed | No public weights listed |
| Released | September 22, 2026 | October 6, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.10 | $1.36 |
| Output $/Mtok | $0.50 | $4.18 |
| Coding index | 57.3 | 56.4 |
| Independent measurements behind it | 5 | 5 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-6 Luna better than Mistral Large 4 for coding?
GPT-6 Luna and Mistral Large 4 are level for coding, 57.3 against 56.4 on the Vibe Coding Index, so price and context decide. Pick Mistral Large 4 for work measured by Terminal-Bench 4 (Vals AI) or LMArena Text: coding, where it leads. Pick GPT-6 Luna for output tokens 8.4x cheaper ($0.50 vs $4.18 per million) and a larger context window (1.05M against 1.05M tokens).
Which is cheaper, GPT-6 Luna or Mistral Large 4?
GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens; Mistral Large 4 costs $1.36 and $4.18. Cost per solved task is known for neither of them.
Which has the bigger context window, GPT-6 Luna or Mistral Large 4?
GPT-6 Luna, with 1.05M tokens against 1.05M.
How many benchmarks is the GPT-6 Luna vs Mistral Large 4 comparison based on?
79 benchmarks have a result for both models as of October 22, 2026. GPT-6 Luna's coding index rests on 5 independent benchmarks and Mistral Large 4's on 5. Where only one model has a result, the page says so instead of naming a winner.