Benchmark Almanac

Grok 4.1

Grok 4.1.

xAI Released 17 Nov 2025 Family Grok 4.1 The lab's model card
20
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
xAI Nov 2025

Grok 4.1

Closed weights Flagship

No Epoch Capabilities Index score yet
  • Virology Capabilities Test (VCT) 61% #4 /28
  • LAB-Bench FigQA 34% #10 /10
  • BioLP-bench 37% #5 /9

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

BenchmarkScoreRankSource
LMArena (Elo) 1,465 #46 of 333 Official leaderboard
EQ-Bench 3 1,586 #1 of 1 Model card

Math and science

BenchmarkScoreRankSource
ResearchClawBench 13.5 #20 of 23 Official leaderboard

Coding

BenchmarkScoreRankSource
Code Arena 1,214 #118 of 124 Official leaderboard
WebDev Arena 1,211 #102 of 105 Epoch AI

Professional work

BenchmarkScoreRankSource
prinzbench 25.0 #18 of 26 Official leaderboard

Safety and security

BenchmarkScoreRankSource
Virology Capabilities Test (VCT) 61% #4 of 28 Model card
Cybench 39% #9 of 24 Official leaderboard
LAB-Bench CloningScenarios 46% #6 of 11 Model card
BioLP-bench 37% #5 of 9 Model card
AgentDojo 5.00% #4 of 7 Model card
LAB-Bench ProtocolQA 79% #9 of 15 Model card
MASK 49% #15 of 21 Model card
AgentHarm 14% #5 of 6 Model card
LAB-Bench FigQA 34% #10 of 10 Model card
MakeMeSay 0% #9 of 9 Model card
General refusals (xAI) 7.00% #7 of 7 Model card
Refusals + System Jailbreak (xAI) 2.00% #4 of 4 Model card
Refusals + User Jailbreak (xAI) 2.00% #4 of 4 Model card
SycophancyEval (answer sycophancy) 19% #3 of 3 Model card

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. 15 of them aren't in the launch post.