Introducing Grok 4.6 12 Aug 2026
- AA Intelligence Index 61
- GDPVal-AA v2 1753
- CursorBench v3.2 69.9
- DeepSWE v1.1 65.9
- FrontierCode v1.1 (Extended) 61.3
- APEX-Agents 57.5
- Terminal-Bench v3.0 26
- APEX-SWE 56.4
- AA-Briefcase 1577
- Harvey LAB (Vals) 15.8
Grok 4.6
Grok 4.6 ranks first on 3 of the 154 benchmarks it has scores on.
178
Benchmarks with a score
3
Ranked first, out of five or more models
6
In the top three
xAI Aug 2026
Grok 4.6
Closed weights Flagship
156.9
#27 of 271
Epoch Capabilities Index
- DeepSWE 67.5% #9 /28
- Terminal-Bench 4.0 20.3% #16 /23
- ZeroBench 17% #18 /66
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BullshitBench | 63.6% | #15 of 107 | Official leaderboard | |
| LMArena (Elo) | 1,455 | #57 of 333 | Official leaderboard | |
| BullshitBench V2 | 66% | #24 of 122 | Official leaderboard | |
| SimpleQA Verified | 52.9% | #15 of 56 | Official leaderboard | |
| Extended NYT Connections | 80% | #38 of 97 | Official leaderboard | |
| xAI Factuality (Hallucination) eval | 1.70% | #1 of 2 | Model card | |
| Document Arena | 1,461 | #20 of 37 | Official leaderboard | |
| FACTS Parametric V2 | 44.5% | #15 of 20 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| DTBench | 95.5% | #5 of 178 | Official leaderboard | |
| GPQA Diamond | 94% | #13 of 219 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 99.2% | #18 of 202 | Epoch AI | |
| SimpleBench | 75.9% | #14 of 100 | Official leaderboard | |
| CritPt | 19.7% | #31 of 160 | Epoch AI | |
| BixBench3 | 48.8 | #5 of 20 | Official leaderboard | |
| ARC-AGI-2 | 67.1% | #27 of 95 | Official leaderboard | |
| ProofBench | 45% | #17 of 59 | Official leaderboard | |
| LatchBio Capabilities v1.0 | 47.4% | #8 of 27 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 66% | #35 of 89 | Epoch AI | |
| ARC-AGI | 87.5% | #38 of 92 | Official leaderboard | |
| MysteryMechanism | 30.6% | #13 of 29 | Official leaderboard | |
| FrontierMath Tier 4 v2 | 31.7% | #32 of 71 | Epoch AI | |
| Prophet Arena | 85.4 | #9 of 18 | Official leaderboard | |
| ExplorationBench | 69% | #5 of 10 | Official leaderboard | |
| BixBench | 93.8% | #2 of 4 | Model card | |
| ProofBench v1.1 | 51% | #29 of 51 | Official leaderboard | |
| xbench-ScienceQA | 70.4% | #19 of 29 | Official leaderboard | |
| SciUniverse | 26.2% | #4 of 6 | Official leaderboard | |
| FrontierPhysics | 3.00% | #13 of 19 | Official leaderboard | |
| Terminal-Bench-Science | 7.14% | #14 of 17 | Official leaderboard | |
| ForecastBench | 65.5 | #77 of 86 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| WebDev Arena | 1,618 | #15 of 105 | Epoch AI | |
| FrontierCode 1.1 | 61.3% | #2 of 12 | Model card | |
| Code Arena | 1,616 | #25 of 124 | Official leaderboard | |
| WeirdML | 67.3% | #23 of 112 | Official leaderboard | |
| ALE-Bench | 1,508 | #25 of 121 | Epoch AI | |
| FrontierCode | 48% | #10 of 46 | Epoch AI | |
| Image-to-WebDev Arena | 1,603 | #14 of 55 | Official leaderboard | |
| Code Migration | 44.6% | #20 of 78 | Official leaderboard | |
| Vals RSI Index | 0.26 | #8 of 25 | Official leaderboard | |
| DeepSWE | 67.5% | #9 of 28 | Official leaderboard | |
| Vibe Code Bench | 76.2% | #37 of 115 | Official leaderboard | |
| APEX-SWE | 56.4% | #1 of 2 | Model card | |
| AutoResearchExam | 0.49 | #5 of 9 | Official leaderboard | |
| Senior SWE-Bench | 26.3% | #13 of 23 | Official leaderboard | |
| MirrorCode | 41.1% | #16 of 28 | Epoch AI | |
| IOI v2 (Vals) | 47.6% | #29 of 45 | Official leaderboard | |
| FrontierSWE | 25.3% | #14 of 21 | Epoch AI | |
| RSI-Exam | 36.7% | #12 of 18 | Official leaderboard | |
| CursorBench | 41.4% | #13 of 19 | Epoch AI | |
| SWE-Marathon | 31.9% | #8 of 11 | Model card | |
| Vibe Code Bench 1-100 | 14.8% | #20 of 26 | Official leaderboard | |
| Brokk Power Ranking | 1.40 | #19 of 19 | Official leaderboard | |
| CursorBench 3.2 | 70.8% | #2 of 2 | Model card | |
| InferenceEval | 46.9% | #1 of 1 | Model card | |
| KernelBenchInternal | 37.2% | #1 of 1 | Model card | |
| SpaceXAI MTS Eval | 61.1% | #1 of 1 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Game Arena Ultimate Tic-Tac-Toe | 505 | #4 of 36 | Official leaderboard | |
| Game Arena Lines of Action | 657 | #9 of 36 | Official leaderboard | |
| Game Arena Four in a Row | 493 | #11 of 38 | Official leaderboard | |
| Game Arena Dark Hex | 288 | #12 of 38 | Official leaderboard | |
| CEO-Bench | $246,419 | #6 of 19 | Official leaderboard | |
| Mystery Game Puzzles | 34% | #26 of 80 | Epoch AI | |
| Game Arena Bargaining | 218 | #12 of 36 | Official leaderboard | |
| LLM Chess | 443 | #43 of 124 | Official leaderboard | |
| Game Arena Go | 1,016 | #13 of 36 | Official leaderboard | |
| Game Arena Dots and Boxes | 948 | #14 of 37 | Official leaderboard | |
| RecreationBench | 36.7% | #4 of 10 | Official leaderboard | |
| Game Arena Nine Men's Morris | 678 | #15 of 36 | Official leaderboard | |
| Game Arena Word Association | 267 | #16 of 36 | Official leaderboard | |
| Terminal-Bench 3.0 | 26.5% | #5 of 11 | Official leaderboard | |
| Game Arena Coin Game | 368 | #16 of 35 | Official leaderboard | |
| Game Arena Chess | 808 | #18 of 39 | Official leaderboard | |
| Game Arena Poker | 0.18 | #15 of 31 | Official leaderboard | |
| Kaggle Game Arena | 316 | #19 of 39 | Official leaderboard | |
| EBR-bench | 30.5% | #14 of 28 | Epoch AI | |
| ARC-AGI-3 | 2.11% | #12 of 24 | Official leaderboard | |
| Agent Arena | 0.00 | #29 of 53 | Official leaderboard | |
| Chess Puzzles | 33% | #46 of 83 | Epoch AI | |
| Game Arena Word Art | 328 | #20 of 35 | Official leaderboard | |
| DeepSearchQA | 81.6% | #13 of 22 | Model card | |
| Game Arena Reversi | 358 | #22 of 36 | Official leaderboard | |
| Game Arena Clobber | 156 | #24 of 36 | Official leaderboard | |
| Terminal-Bench 4.0 | 20.3% | #16 of 23 | Official leaderboard | |
| ApprenticeBench | 13% | #13 of 18 | Official leaderboard | |
| Game Arena Checkers | 553 | #31 of 38 | Official leaderboard | |
| Time Horizon Index: KSP | 5.83% | #12 of 13 | Official leaderboard | |
| Long-Horizon-Terminal-Bench 2.0 | 42% | #14 of 15 | Official leaderboard | |
| Brood War Bench | 11.1% | #6 of 6 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OR-License | 88.4% | #1 of 10 | Official leaderboard | |
| LegalBench | 86.3% | #14 of 138 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 15.8% | #9 of 80 | Official leaderboard | |
| Legal Research Bench | 48.1% | #12 of 88 | Official leaderboard | |
| CorpFin v2 | 66.2% | #21 of 119 | Official leaderboard | |
| APEX-Agents | 65.3% | #10 of 56 | Epoch AI | |
| Tax Agent Bench | 33.1% | #16 of 79 | Official leaderboard | |
| Vending-Bench 2 | $9,047 | #14 of 66 | Official leaderboard | |
| MedScribe | 86.5% | #22 of 98 | Official leaderboard | |
| GDPval-AA v2 | 1,753 | #6 of 26 | Model card | |
| Public Benefits Bench | 66.8% | #17 of 50 | Official leaderboard | |
| Clipping Bench | $805 | #7 of 19 | Official leaderboard | |
| EEBench | 57.1% | #10 of 26 | Official leaderboard | |
| MedCode | 44.7% | #38 of 96 | Official leaderboard | |
| Vals Finance Agent v2 | 53.7% | #32 of 80 | Official leaderboard | |
| PolicyBench | 85.8% | #20 of 50 | Official leaderboard | |
| Excel Modeling Benchmark | 62.7% | #31 of 76 | Official leaderboard | |
| PharmaBench (micro1) | 12.5% | #6 of 12 | Official leaderboard | |
| FoodTruck Bench | $59,537 | #5 of 10 | Official leaderboard | |
| AA-Briefcase | 1,577 | #3 of 6 | Model card | |
| Epoch Automation Reports | 59.1% | #3 of 6 | Official leaderboard | |
| 3DCodeBench | 54% | #1 of 2 | Model card | |
| MortgageTax | 64.2% | #44 of 86 | Official leaderboard | |
| CAD Arena | 0.35 | #7 of 12 | Official leaderboard | |
| TaxEval v2 | 71.1% | #79 of 129 | Official leaderboard | |
| BlenderWorld | 57.0 | #8 of 13 | Official leaderboard | |
| Immigration Bench | 66% | #15 of 24 | Official leaderboard | |
| RealmPathology | 72.1% | #9 of 13 | Official leaderboard | |
| Parametric CAD Bench | 27.7% | #8 of 11 | Official leaderboard | |
| OfficeQA | 63.2% | #10 of 12 | Model card | |
| micro1 Medical Index | 64.1% | #7 of 8 | Official leaderboard | |
| CADGenBench | 40.9% | #2 of 2 | Model card | |
| PartBench | 58.5% | #1 of 1 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| VoxelBench | 2,023 | #12 of 56 | Official leaderboard | |
| Vision Arena | 1,262 | #36 of 136 | Official leaderboard | |
| ZeroBench | 17% | #18 of 66 | Official leaderboard | |
| Blueprint-Bench 2 | 33.2% | #14 of 35 | Official leaderboard | |
| ArtifactArena | 1,136 | #9 of 21 | Official leaderboard | |
| ClockBench | 20% | #19 of 37 | Official leaderboard | |
| Furniture Assembly Benchmark | 40% | #18 of 35 | Epoch AI | |
| GDP.pdf | 17.2% | #26 of 39 | Epoch AI | |
| Code2Video Benchmark | 1,450 | #17 of 19 | Official leaderboard | |
| 1H-VideoQA | 60% | #18 of 20 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Virology Capabilities Test (VCT) | 67.4% | #1 of 28 | Model card | |
| TxBench-Antibody-Discovery | 53% | #6 of 28 | Official leaderboard | |
| TxBench-Preclinical-Pharmacology | 61.7% | #6 of 28 | Official leaderboard | |
| BioSecBench-Refusal | 46.9% | #6 of 22 | Official leaderboard | |
| TxBench-Oligo-Discovery | 46% | #8 of 28 | Official leaderboard | |
| BioSecBench-Function | 44% | #8 of 27 | Official leaderboard | |
| CyberGym | 79.7% | #11 of 33 | Official leaderboard | |
| BioSecBench-Surveillance | 48% | #9 of 27 | Official leaderboard | |
| Autointent-Bio | 100% | #1 of 3 | Model card | |
| Autointent-Chem | 100% | #1 of 3 | Model card | |
| LAB-Bench | 80.7% | #1 of 3 | Model card | |
| xAI sycophancy eval | 0.04% | #1 of 3 | Model card | |
| Human Pathogen Capabilities Test v2 (HPCT v2) | 62.1% | #24 of 71 | Official leaderboard | |
| VariantBench | 40.4% | #10 of 27 | Official leaderboard | |
| MetagenomicsBench | 42% | #11 of 26 | Official leaderboard | |
| OncologyBench | 18.7% | #13 of 30 | Official leaderboard | |
| EpiBench | 23.6% | #12 of 27 | Official leaderboard | |
| Biosecurity VCT | 47.8% | #1 of 2 | Model card | |
| scBench-Long | 16.5% | #15 of 27 | Official leaderboard | |
| General refusals (xAI) | 0.93% | #4 of 7 | Model card | |
| CyberBench v1.1 (Vals) | 67.7% | #25 of 43 | Official leaderboard | |
| CWE-bench | 57% | #7 of 12 | Official leaderboard | |
| SpatialBench-Long | 27.4% | #16 of 27 | Official leaderboard | |
| Child safety (xAI) | 0% | #2 of 3 | Model card | |
| FORTRESS-RN | 97.9% | #2 of 3 | Model card | |
| HackerBench | 6.90% | #2 of 3 | Model card | |
| ProtocolQA Open-Ended (xAI) | 79.6% | #2 of 3 | Model card | |
| Self-harm refusals (xAI) | 0.84% | #2 of 3 | Model card | |
| Standard jailbreaks (xAI) | 0.04% | #2 of 3 | Model card | |
| DeepsecBench | 16.1 | #19 of 28 | Official leaderboard | |
| CVE-Bench | 39.8% | #12 of 14 | Model card | |
| StrongREJECT | 3.90% | #19 of 22 | Model card | |
| PCT (Polarized Contrastive Pairs) | 55.4% | #40 of 46 | Official leaderboard | |
| MASK | 1.90% | #19 of 21 | Model card | |
| WellnessBench | 43% | #12 of 12 | Official leaderboard | |
| BioUseBench | 90.7% | #2 of 2 | Model card | |
| Long-horizon jailbreaks (xAI) | 1.00% | #2 of 2 | Model card | |
| SecureCodeReview (xAI) | 58.7% | #1 of 1 | Model card |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Artificial Analysis Intelligence Index v4.1 | 61.0 | #1 of 7 | Model card | |
| Conceptual Reasoning Index | 62.5 | #30 of 149 | Official leaderboard | |
| LMCA | 48.5% | #30 of 139 | Epoch AI | |
| CAIS Text Capabilities Index | 43.5 | #15 of 66 | Official leaderboard | |
| ACCoRD | 71.6 | #38 of 150 | Official leaderboard | |
| CAIS Risk Index | 48.8 | #26 of 65 | Official leaderboard | |
| LiveBench | 67.8% | #15 of 28 | Official leaderboard | |
| Vals Science Index | 38.1% | #13 of 23 | Official leaderboard | |
| SAGE (Vals) | 28.9% | #78 of 81 | Official leaderboard |
xAI's models over time
xAI's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 34 of them aren't in the launch post.
Model Card: Grok 4.6 12 Aug 2026
Capabilities
- CursorBench 3.2 70.8
- APEX-SWE 56.4
- FrontierCode 1.1 61.3
- DeepSWE 67
- SWE-Marathon 31.9
- Terminal-Bench 3.0 26
- GDPval-AA v2 1753
- AA-Briefcase 1577
- APEX-Agents 57.5
- OfficeQA 63.2
- Harvey LAB (Legal Agent Benchmark) 15.8
- EEBench 60
- 3DCodeBench 54
- PartBench 58.5
- CADGenBench 40.9
- Parametric CAD Bench 88.4
- DeepSearchQA 81.6
Hallucination
Safety
- General refusals (xAI) 0.93
- Child safety (xAI) 0
- Autointent-Bio 100
- Autointent-Chem 100
- FORTRESS-RN 97.9
- Self-harm refusals (xAI) 0.84
Robustness
Alignment
- MASK 1.9
- xAI sycophancy eval 0.04
Bio and chem risk
- Virology Capabilities Test (VCT) 67.4
- Biosecurity VCT 47.8
- BioUseBench 90.7
- WMDP
- LAB-Bench 80.7
- ProtocolQA Open-Ended (xAI) 79.6
- BixBench 93.8
Cyber risk
- CyberGym 79.7
- CVE-Bench 39.8
- SecureCodeReview (xAI) 58.7
- HackerBench 6.9
AI R&D risk
- SpaceXAI MTS Eval 61.1
- InferenceEval 46.9
- KernelBenchInternal 37.2