Grok 4.6 vs Mistral Large
Updated October 21, 2026. 37 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
vs boards won of 37 shared
Decision reading
Only Grok 4.6 has a coding score, 65.7 on the Vibe Coding Index; Mistral Large has none, so the two cannot be ranked on coding.
Compare them on price, context and the boards below.
Model AGrok 4.6xAI65.7/1006 independent benchmarks$0.82 per solved taskModel BMistral LargeMistral AINot benchmarkedNo coding benchmarkNo cost per solved task
- Shared results
- 37
- Grok 4.6 only
- 86
- Mistral Large only
- 7
- Like-for-like categories
- 6 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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 Grok 4.6 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Grok 4.6
Grok 4.6 leads, 44.3 against 5.8.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Grok 4.6 has a AA Agentic Index result, 53.0.
Directional only
Reasoning
GPQA Diamond
Grok 4.6
Grok 4.6 leads, 94.7% against 47.7%.
Same board, both measured
Math
AIME
No call
Only Mistral Large has a AIME result, 9.2%.
Directional only
Knowledge
MMLU Pro
Grok 4.6
Grok 4.6 leads, 89.4% against 69.7%.
Same board, both measured
Web design
Design Arena: websites
No call
Only Grok 4.6 has a Design Arena: websites result, 1289.
Directional only
Long documents
Listed context window
Grok 4.6
Grok 4.6 takes 500K tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
No call
Grok 4.6 $0.0050, Mistral Large $0.0050 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Grok 4.6 has the number, $0.82.
Directional only
At a glance
- Coding index
- Grok 4.6 65.7 from 6 benchmarks, Mistral Large not benchmarked.
- Widest category
- Reasoning, 42.0 points apart on average over 6 shared benchmarks.
- Widest single gap
- GPQA Diamond (Artificial Analysis run): Grok 4.6 94.9% against Mistral Large 35.1%, 59.8 points apart on a 0 to 100 scale.
- Output price
- Both cost $6.00 per million output tokens.
- Context window
- Grok 4.6 takes 500K tokens, Mistral Large 128K.
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 | Grok 4.6 | Bars | Mistral Large | Ties |
|---|---|---|---|---|
| Coding | 3 | 0 | ||
| Intelligence | 2 | 0 | ||
| Reasoning | 6 | 0 | ||
| Math | 2 | 0 | ||
| Knowledge | 6 | 0 | ||
| General | 18 | 0 | ||
| All shared boards | 37 | 0 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| LiveBench coding | 89.1% | Not measured | Only Grok 4.6 measured |
| LMArena WebDev | 1617 | Not measured | Only Grok 4.6 measured |
| AA Coding Index | 76.8 | Not measured | Only Grok 4.6 measured |
| LiveCodeBench (Vals AI) | 88.2% | 37.1% | Grok 4.6 |
| SWE-bench Verified (Vals AI) | 95.6% | Not measured | Only Grok 4.6 measured |
| Terminal-Bench 4 (Vals AI) | 17.2% | Not measured | Only Grok 4.6 measured |
| Vibe Code Bench (Vals AI) | 76.2% | Not measured | Only Grok 4.6 measured |
| Code migration (Vals AI) | 44.6% | Not measured | Only Grok 4.6 measured |
| ProgramBench (Vals AI) | 1.0% | Not measured | Only Grok 4.6 measured |
| IOI (Vals AI) | 47.6% | Not measured | Only Grok 4.6 measured |
| Terminal-Bench 2.1 (Vals AI) | 78.3% | Not measured | Only Grok 4.6 measured |
| Vibe Code Bench 1-100 (Vals AI) | 14.8% | Not measured | Only Grok 4.6 measured |
| SkillsBench (Vals AI) | 55.8% | Not measured | Only Grok 4.6 measured |
| CyberBench (Vals AI) | 67.7% | Not measured | Only Grok 4.6 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 17.8% | Only Mistral Large measured |
| SciCode | 56.5% | Not measured | Only Grok 4.6 measured |
| FrontierCode | 48.0% | Not measured | Only Grok 4.6 measured |
| FrontierSWE | 25.3% | Not measured | Only Grok 4.6 measured |
| WeirdML | 67.3% | Not measured | Only Grok 4.6 measured |
| ALE-Bench | 1508.1 | Not measured | Only Grok 4.6 measured |
| LMArena Text: coding | 1466 | 1275 | Grok 4.6 |
| LMArena Text: software and IT | 1457 | 1274 | Grok 4.6 |
| LMArena WebDev: React | 1631 | Not measured | Only Grok 4.6 measured |
| LMArena WebDev: HTML | 1597 | Not measured | Only Grok 4.6 measured |
| LMArena WebDev: from an image | 1603 | Not measured | Only Grok 4.6 measured |
| Neither model has a result on the other 13 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| AA Intelligence Index | 44.3 | 5.8 | Grok 4.6 |
| Epoch Capabilities Index | 156.6 | 122.1 | Grok 4.6 |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index22 boards, none shared
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| AA Agentic Index | 53.0 | Not measured | Only Grok 4.6 measured |
| Design Arena: agents building web apps | 1248 | Not measured | Only Grok 4.6 measured |
| Design Arena: agents building full-stack apps | 1264 | Not measured | Only Grok 4.6 measured |
| Design Arena: agents building mobile apps | 1236 | Not measured | Only Grok 4.6 measured |
| Design Arena: agents building Android apps | 1297 | Not measured | Only Grok 4.6 measured |
| Design Arena: agents building Godot games | 1276 | Not measured | Only Grok 4.6 measured |
| Design Arena: agents building games | 1208 | Not measured | Only Grok 4.6 measured |
| Design Arena: HTML slides | 1249 | Not measured | Only Grok 4.6 measured |
| Design Arena: PowerPoint slides | 1241 | Not measured | Only Grok 4.6 measured |
| LiveBench research | 84.5% | Not measured | Only Grok 4.6 measured |
| LiveBench browser | 64.0% | Not measured | Only Grok 4.6 measured |
| LiveBench long agentic | 40.7% | Not measured | Only Grok 4.6 measured |
| Finance Agent (Vals AI) | 53.7% | Not measured | Only Grok 4.6 measured |
| Tax Agent Bench (Vals AI) | 33.1% | Not measured | Only Grok 4.6 measured |
| Legal Research Bench (Vals AI) | 48.1% | Not measured | Only Grok 4.6 measured |
| Excel Modeling Benchmark (Vals AI) | 62.7% | Not measured | Only Grok 4.6 measured |
| BioMysteryBench (Vals AI) | 72.2% | Not measured | Only Grok 4.6 measured |
| Terminal-Bench-Science (Vals AI) | 11.4% | Not measured | Only Grok 4.6 measured |
| Vals Science Index | 38.1% | Not measured | Only Grok 4.6 measured |
| Time Horizon Index (Vals AI) | 5.8% | Not measured | Only Grok 4.6 measured |
| APEX-Agents | 65.3% | Not measured | Only Grok 4.6 measured |
| Vending-Bench 2 | $9,047 | Not measured | Only Grok 4.6 measured |
| Neither model has a result on the other 17 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| GPQA Diamond | 94.7% | 47.7% | Grok 4.6 |
| LiveBench reasoning | 91.2% | Not measured | Only Grok 4.6 measured |
| MysteryMechanism (Vals AI) | 30.6% | Not measured | Only Grok 4.6 measured |
| GPQA Diamond (Artificial Analysis run) | 94.9% | 35.1% | Grok 4.6 |
| Humanity's Last Exam (Artificial Analysis run) | 44.1% | 3.5% | Grok 4.6 |
| GPQA Diamond (Epoch AI run) | 94.0% | Not measured | Only Grok 4.6 measured |
| ARC-AGI | 87.5% | Not measured | Only Grok 4.6 measured |
| ARC-AGI-2 | 67.1% | Not measured | Only Grok 4.6 measured |
| SimpleBench | 75.9% | Not measured | Only Grok 4.6 measured |
| Chess puzzles | 33.0% | Not measured | Only Grok 4.6 measured |
| Mystery game puzzles | 34.0% | Not measured | Only Grok 4.6 measured |
| EBR-bench | 30.5% | Not measured | Only Grok 4.6 measured |
| LMArena Text: hard prompts | 1448 | 1257 | Grok 4.6 |
| LMArena Text: expert | 1469 | 1208 | Grok 4.6 |
| LMArena Text: hard prompts in English | 1445 | 1268 | Grok 4.6 |
| LMArena Vision: diagrams | 1285 | Not measured | Only Grok 4.6 measured |
| LMArena Vision: homework | 1283 | Not measured | Only Grok 4.6 measured |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| AIME | Not measured | 9.2% | Only Mistral Large measured |
| MATH 500 (Vals AI) | Not measured | 74.4% | Only Mistral Large measured |
| MGSM (Vals AI) | Not measured | 87.2% | Only Mistral Large measured |
| OTIS Mock AIME | 99.2% | Not measured | Only Grok 4.6 measured |
| FrontierMath, tiers 1 to 3 | 66.0% | Not measured | Only Grok 4.6 measured |
| FrontierMath, tier 4 | 31.7% | Not measured | Only Grok 4.6 measured |
| ProofBench | 51.0% | Not measured | Only Grok 4.6 measured |
| LMArena Text: math | 1417 | 1261 | Grok 4.6 |
| LMArena Text: mathematical work | 1464 | 1257 | Grok 4.6 |
| Neither model has a result on the other 4 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| MMLU Pro | 89.4% | 69.7% | Grok 4.6 |
| LegalBench (Vals AI) | 86.3% | Not measured | Only Grok 4.6 measured |
| CorpFin (Vals AI) | 66.2% | Not measured | Only Grok 4.6 measured |
| TaxEval (Vals AI) | 71.1% | 63.8% | Grok 4.6 |
| MortgageTax (Vals AI) | 64.2% | Not measured | Only Grok 4.6 measured |
| MedQA (Vals AI) | Not measured | 76.2% | Only Mistral Large measured |
| MedCode (Vals AI) | 44.7% | Not measured | Only Grok 4.6 measured |
| MMLU Pro (Artificial Analysis run) | Not measured | 51.5% | Only Mistral Large measured |
| SimpleQA Verified | 49.3% | Not measured | Only Grok 4.6 measured |
| MMLU | Not measured | 68.8% | Only Mistral Large measured |
| LMArena Text: science | 1437 | 1260 | Grok 4.6 |
| LMArena Text: medicine and healthcare | 1420 | 1255 | Grok 4.6 |
| LMArena Text: legal and government | 1429 | 1275 | Grok 4.6 |
| LMArena Text: business and finance | 1416 | 1238 | Grok 4.6 |
| Neither model has a result on the other 3 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| Design Arena: websites | 1289 | Not measured | Only Grok 4.6 measured |
| Design Arena: UI components | 1295 | Not measured | Only Grok 4.6 measured |
| Design Arena: game dev | 1296 | Not measured | Only Grok 4.6 measured |
| Design Arena: data viz | 1282 | Not measured | Only Grok 4.6 measured |
| Design Arena: 3D | 1277 | Not measured | Only Grok 4.6 measured |
| Design Arena: all code categories | 1292 | Not measured | Only Grok 4.6 measured |
| Design Arena: SVG | 1239 | Not measured | Only Grok 4.6 measured |
| Design Arena: ASCII art | 1278 | Not measured | Only Grok 4.6 measured |
| LiveBench design | 79.8% | Not measured | Only Grok 4.6 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index180
| Benchmark | Grok 4.6 | Mistral Large | Leader |
|---|---|---|---|
| LiveBench data analysis | 75.3% | Not measured | Only Grok 4.6 measured |
| Public Benefits Bench (Vals AI) | 66.8% | Not measured | Only Grok 4.6 measured |
| SAGE (Vals AI) | 28.9% | Not measured | Only Grok 4.6 measured |
| MedScribe (Vals AI) | 86.5% | Not measured | Only Grok 4.6 measured |
| Long context reasoning (AA-LCR) | 81.0% | Not measured | Only Grok 4.6 measured |
| DTBench | 97.3% | 60.8% | Grok 4.6 |
| LMCA | 48.5% | Not measured | Only Grok 4.6 measured |
| Furniture assembly | 40.0% | Not measured | Only Grok 4.6 measured |
| LMArena Text | 1429 | 1265 | Grok 4.6 |
| LMArena Text: instruction following | 1431 | 1249 | Grok 4.6 |
| LMArena Text: creative writing | 1426 | 1242 | Grok 4.6 |
| LMArena Vision | 1263 | Not measured | Only Grok 4.6 measured |
| LMArena Text: English | 1433 | 1284 | Grok 4.6 |
| LMArena Text: non-English | 1420 | 1235 | Grok 4.6 |
| LMArena Text: Chinese | 1476 | 1239 | Grok 4.6 |
| LMArena Text: German | 1426 | 1242 | Grok 4.6 |
| LMArena Text: Japanese | 1378 | 1170 | Grok 4.6 |
| LMArena Text: Korean | 1391 | 1202 | Grok 4.6 |
| LMArena Text: French | 1455 | 1325 | Grok 4.6 |
| LMArena Text: Spanish | 1414 | 1268 | Grok 4.6 |
| LMArena Text: Russian | 1421 | 1252 | Grok 4.6 |
| LMArena Text: Polish | 1430 | 1157 | Grok 4.6 |
| LMArena Text: multi-turn | 1425 | 1260 | Grok 4.6 |
| LMArena Text: longer queries | 1456 | 1261 | Grok 4.6 |
| LMArena Text: writing and language | 1429 | 1261 | Grok 4.6 |
| LMArena Text: entertainment and media | 1421 | 1230 | Grok 4.6 |
| LMArena Vision: English | 1270 | Not measured | Only Grok 4.6 measured |
| LMArena Vision: Chinese | 1289 | Not measured | Only Grok 4.6 measured |
| LMArena Vision: reading text | 1272 | Not measured | Only Grok 4.6 measured |
| LMArena Vision: humor | 1270 | Not measured | Only Grok 4.6 measured |
| LMArena Vision: creative writing | 1262 | Not measured | Only Grok 4.6 measured |
| LMArena Document | 1452 | Not measured | Only Grok 4.6 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 | Grok 4.6 | Mistral Large | 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
Grok 4.6$0.0050Fits in one request
Mistral Large$0.0050Fits in one request
Same cost
Repository review
50K input + 3K output tokens
Grok 4.6$0.118Fits in one request
Mistral Large$0.118Fits in one request
Same cost
Agent loop
200K input + 10K output tokens
Grok 4.6$0.460Fits in one request
Mistral Large$0.460Exceeds the context window
Same cost
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 | SpaceXAI Grok 4.6 | Mistral Large |
|---|---|---|
| Maker | xAI | Mistral AI |
| Weights | No public weights listed | No public weights listed |
| Released | August 12, 2026 | February 26, 2024 |
| Context window | 500K | 128K |
| Input $/Mtok | $2.00 | $2.00 |
| Output $/Mtok | $6.00 | $6.00 |
| Coding index | 65.7 | Not benchmarked |
| Independent measurements behind it | 6 | 0 |
| Cost per solved task | $0.82 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Grok 4.6 better than Mistral Large for coding?
Only Grok 4.6 has a coding score, 65.7 on the Vibe Coding Index; Mistral Large has none, so the two cannot be ranked on coding. Compare them on price, context and the boards below.
Which is cheaper, Grok 4.6 or Mistral Large?
Grok 4.6 costs $2.00 per million input tokens and $6.00 per million output tokens; Mistral Large costs $2.00 and $6.00. Cost per solved task is known for Grok 4.6 only.
Which has the bigger context window, Grok 4.6 or Mistral Large?
Grok 4.6, with 500K tokens against 128K.
How many benchmarks is the Grok 4.6 vs Mistral Large comparison based on?
37 benchmarks have a result for both models as of October 21, 2026. Grok 4.6's coding index rests on 6 independent benchmarks and Mistral Large's on 0. Where only one model has a result, the page says so instead of naming a winner.