AI Battle > Rankings > Grok 4.6 (high)

Grok 4.6 (high) — Placements in 61 Rankings (2026)

Grok 4.6 (high), published by SpaceXAI, appears in 61 of the 129 datasets AI Battle tracks. Its strongest published placement is #2 of 20 in 𝜏³-Banking: Score.

Last updated: . Every placement below comes from the named public dataset; AI Battle joins them by slug and changes nothing else.

All placements for Grok 4.6 (high)

RankingRankValue
𝜏³-Banking: Score#2 of 200.51
AA-Briefcase Elo#5 of 201534
AA-Omniscience Index#5 of 2030.5
AA-Omniscience Index#5 of 2030.5
AA-Briefcase Elo#5 of 201534
SciCode: Output Tokens per Task#5 of 20352
GDPval-AA v2 Leaderboard#6 of 201643
Artificial Analysis Finance & Accounting Index#6 of 2049.5
AA-Omniscience Index: Cost Breakdown#7 of 201.77
Artificial Analysis Intelligence Index by Open Weights / Proprietary#8 of 2044.4
Artificial Analysis Intelligence Index by Open Weights / Proprietary#8 of 2044.4
Artificial Analysis Intelligence Index#8 of 2044.4
Artificial Analysis Intelligence Index#8 of 2044.4
AA-Omniscience Hallucination Rate#8 of 200.34
Output Tokens per Intelligence Index Task#9 of 2017053
Output Tokens per Intelligence Index Task#9 of 2017053
AA-Omniscience Index: Token Usage#9 of 20887417
Humanity's Last Exam: Cost per Task#10 of 200.00
Humanity's Last Exam: Output Tokens per Task#10 of 20250
AA-Omniscience Index: Score#10 of 2030.5
AA-Briefcase Rubric Pass Rate by File Type (Normalized)#10 of 200.82
AA-Briefcase Elo#10 of 201534
AA-Briefcase Output Tokens per Task#10 of 2061226
SciCode: Cost per Task#10 of 200.00
Cost per Task#6 of 111.86
Intelligence#6 of 1144.4
AA-AnalystAgent pass^5#6 of 110.41
Cost per Task#6 of 111.86
Intelligence#6 of 1144.4
GDPval-AA v2 Leaderboard#11 of 201643
Terminal-Bench v4.0: Output Tokens per Task#12 of 2036898
AA-Omniscience Accuracy#13 of 200.48
GDPval-AA v2: Output Tokens per Task#13 of 2036002
End-to-End Response Time#14 of 2038.6
Latency: Time To First Answer Token#14 of 2038.6
Humanity's Last Exam: Score#14 of 200.43
SciCode: Score#14 of 200.56
Cost to Run Artificial Analysis Intelligence Index#15 of 20129
Cost to Run Artificial Analysis Intelligence Index#15 of 20129
AA-LCR v1.1: Output Tokens per Task#15 of 2080.9
SciCode: Time per Task#15 of 201.07
Cost per Intelligence Index Task#16 of 200.03
Pricing: Cache Hit, Input, and Output#16 of 202.00
Cost per Intelligence Index Task#16 of 200.03
Pricing: Cache Hit, Input, and Output#16 of 202.00
Terminal-Bench v4.0: Score#16 of 200.21
Time per Task#16 of 2030.7
𝜏³-Banking: Output Tokens per Task#16 of 202794
Speed#9 of 1158.5
Speed#9 of 1158.5
Time per Intelligence Index Task#17 of 2010.1
Time per Intelligence Index Task#17 of 2010.1
AA-LCR v1.1: Cost per Task#17 of 200.19
Terminal-Bench v4.0: Time per Task#19 of 2024.9
AA-Briefcase Cost per Task#19 of 200.37
AA-LCR v1.1: Score#19 of 200.80
GDPval-AA v2: Cost per Task#19 of 200.00
Output Speed#20 of 2058.5
Output Speed#20 of 2058.5
Terminal-Bench v4.0: Cost per Task#20 of 200.00
GDPval-AA v2: Average Turns per Task#20 of 2037.0

Price vs intelligence

Grok 4.6 (high)intelligence index →input price / 1M → $4.00
Each dot is a model, joined by AI Battle from two public datasets — the intelligence index ranking and the published input pricing ranking. Every published value is shown as-is, no averaging. The dashed line is the value frontier: no model above or to the right of it offers a lower input price at its intelligence level.

How to read this profile

Each row is a separate measurement in a separate dataset — the position reflects the metric of that ranking, and a higher number is not "better" for every dataset (times and prices are costs). AI Battle deliberately does not average placements into a single score: datasets differ in coverage, so a total would hide more than it shows.

Raw data

Machine-readable profile: /data/models/grok-4-6.json.

Method

Placements, joins and this page are computed by AI Battle from public datasets — see the methodology for sources, limitations and corrections.