Grok 4.6: score and benchmark evidence
Historical release · · xAI
114.0 points · Rank 5 among 19 ranked models · Stability range 102.5–126.3 points.
Rank follows the point estimate; it does not establish statistically significant superiority. The scale is anchored at mean 100, SD 15 in the frozen calibration cohort, not human IQ or a percentage.
One capability score combines evidence across seven equally weighted domains. Related benchmark variants share a family budget, and influence from one evaluator is capped.
Release tli-2026-v5.1-2026-09-06. Methodology and limitations for this release.
This page preserves historical evidence. See the current model record.
Capability domains
Domain scores use calibration-cohort standard-deviation units. Unmeasured domains are shown as unavailable, not zero.
| Domain | Score (SD units) | Observed benchmarks |
|---|---|---|
| Knowledge & reasoning | 0.42 | 4 |
| Coding & software engineering | 0.42 | 2 |
| Agentic tool & computer use | 0.49 | 2 |
| Professional & real-world work | Unavailable | 0 |
| Multimodal & vision | 0.39 | 1 |
| Cybersecurity | 0.27 | 1 |
| Preference & communication | Unavailable | 0 |
Published benchmark evidence
10 results contribute to this score, from 59 collected results and 10 contributing benchmark families. Counts are not independent sample sizes.
Only results with a matched source record, metric and documented configuration contribute. Developer-reported results are eligible when their configuration is clear. Unresolved records remain in the database with their exclusion reason. Archived reviews and live source checks are labelled separately.
The bar estimates score stability under the scoring assumptions: a fixed calibration and comparable benchmark availability. It resamples benchmark families and evaluators while keeping each domain’s evidence strength fixed, then adds a frozen correction learned from missing-evidence tests. Benchmark count and domain coverage both affect that correction. It does not add a second full capability-posterior draw or uncertainty from changing the reference calibration. It is not a guarantee under selective publication, new benchmark domains or future model releases.
Reporting sensitivity: ±20.6 points. Sensitivity to selectively published results. This stress range is not a confidence interval or a probability statement.
GDPval-AA v2 (Elo): 1753 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Terminal-Bench 2.1: 88.4%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 (high); AA published evaluation harness
Metric: pass@1 accuracy, AA harness
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
DeepSWE v1.1: 65.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LMArena (Chatbot Arena) — Text: 1464 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA Intelligence Index v4.1.1 (score): 61 native-points
Not used in the score. Source category not recorded. Review status: excluded.
Held-out composite, never a scoring input
GPQA Diamond: 94.9%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 (high); AA published evaluation harness
Metric: accuracy, AA
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
AA-LCR: 75%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Finance Agent v2 — Vals: 53.68%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Excel Modeling Benchmark — Vals: 62.57%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SciCode: 53.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Humanity's Last Exam (no tools): 42.9%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 (high); AA published evaluation harness
Metric: accuracy, no tools, AA
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
Legal Research Bench — Vals: 48.08%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Harvey LAB (Vals): 15.8%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA-Briefcase Overall Elo: 1577 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MedScribe — Vals: 86.53%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SWE-bench Verified — Vals: 95.6%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LegalBench — Vals: 86.31%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
TaxEval v2 — Vals: 71.1%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Vals Index: 71.1%
Not used in the score. Source category not recorded. Review status: excluded.
Held-out composite, never a scoring input
Public Benefits Bench — Vals: 66.85%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MedCode — Vals: 44.71%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA-Omniscience: 30.5 native-points
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 (high); AA published evaluation harness
Metric: AA Omniscience index -100 to 100
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
MMLU Pro — Vals: 89.4%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LiveBench: 78%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MortgageTax — Vals: 64.19%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SAGE — Vals: 28.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Code Migration †: 44.57%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Terminal-Bench v3.0: 26%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
LiveCodeBench: 88.22%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CyberGym: 79.7%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SWE-Marathon v1.1: 31.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CursorBench v3.2: 69.9%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
APEX-Agents Mean Criteria Passed: 57.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
APEX-SWE (Pass@1, Terminus-2): 56.4%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
ARC-AGI-2: 67.1%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
SimpleQA: 53%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
FrontierCode v1.1 Extended: 61.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
OfficeQA Pro †: 63.2%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
FrontierCode v1.1 Main — Anthropic H2H: 48%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
τ³-Banking: 50.7%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 (high); AA published evaluation harness
Metric: strict task success, AA
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
CritPt: 17.1%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 (high); AA published evaluation harness
Metric: accuracy, AA official grading
Original source · Retrieved 2026-09-06
Current first-party evaluator data matches the recorded score at its published rounding. Exact pinned configuration selected; later value changes are not silently substituted.
Terminal-Bench 4.0: 20.3%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
GDP.pdf: 16%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
ProofBench v1.1 — Vals: 51%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Vibe Code Bench v1.1 — Vals: 76.24%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MineBench: 1931 native-points
Not used in the score. Source category not recorded. Review status: excluded.
Unresolved Pro/base model identities in source catalog; exclude until matched.
ARC-AGI-1: 87.5%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
ARC-AGI-3 (standard harness): 2.11%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
EnterpriseOps-Gym-AA: 48.34%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
MLCR-AA (Medical Long Context Reasoning): 12.22%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
AA-AnalystAgent: 41.25%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Vending-Bench 2: 9047.03 native-points
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
CWE-bench: 38.2%
Contributes to the score. Source category: independent. Review status: reviewed-independent.
Configuration: Grok 4.6 High; high reasoning; grok-cli
Metric: v0 Sept 1 deterministic pass@1, 100 tasks; exploit blocked and existing tests pass
Original source · Retrieved 2026-09-06
Published high effort matches this model configuration. Other models pinned to max/xhigh are not approved from this high-only board.
ZeroBench: 17%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
BioMysteryBench v1 — Vals: 72.222%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
Tax Agent Bench v1 — Vals: 70.788%
Not used in the score. Source category not recorded. Review status: unresolved.
Inherited cell lacks a matched, reviewed source/configuration record.
RuneBench (30m, mean ln(1 + XP/min)): 5.621351 native-points
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: grok46-xhigh
Metric: Arithmetic mean of ln(1 + peakXpRate) over all 16 skills, exactly as the owner’s Heatmap.js default ranking. Data rates are already normalized; do not divide by 200 again.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
VoxelBench (text-to-voxel, Glicko-2): 2099 native-points
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: Grok 4.6 (xHigh)
Metric: Glicko-2 rating from the text-to-voxel board; not win percentage, deviation, image-input mode or MineBench Elo.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
Code Arena WebDev (overall): 1625 native-points
Contributes to the score. Source category: independent. Review status: reviewed-archive.
Configuration: grok-4.6-high
Metric: Published overall WebDev preference rating, not task success percentage or Arena Text score.
Original source · Retrieved 2026-09-06
Matched exact metric/value and the archived benchmark-specific configuration review. Unknown native effort stays explicitly unknown.
Cite this model record
The Latent. “Grok 4.6: score and benchmark evidence.” 2026-09-06. Release tli-2026-v5.1-2026-09-06.
Download release data and definitions (JSON) · Data usage terms