evals.report
BenchmarksLabsCompareRun guidesIn the wild
xAIGrok

Grok 4.1 fast reasoning

xAI · Grok. Released Nov 19, 2025.

Grok 4.1 fast reasoning is a model from xAI in the Grok family, released Nov 19, 2025. evals.report tracks 16 reported Grok 4.1 fast reasoning benchmark scores across Berkeley Function Calling Leaderboard, Artificial Analysis Intelligence Index, MMLU-Pro, AIME 2026, GDPval, LiveCodeBench, SciCode, Global-MMLU, and 8 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.

16 results

Benchmark results 16

Compare this model
BenchmarkCategoryScoreMetricStatusDate
Berkeley Function Calling LeaderboardTool use69.57%accuracyOfficialNov 19, 2025Details
Artificial Analysis Intelligence IndexReasoning38.6IndexUnverifiedNov 19, 2025Details
MMLU-ProReasoning85.4%accuracyVerifiedNov 19, 2025Details
AIME 2026Reasoning94.17%accuracyOfficialNov 19, 2025Details
GDPvalAgents1046EloOfficialNov 19, 2025Details
LiveCodeBenchCoding82.2%Pass@1UnverifiedNov 19, 2025Details
SciCodeCoding44.2%accuracyUnverifiedNov 19, 2025Details
Global-MMLUReasoning85.6%accuracyUnverifiedNov 19, 2025Details
WebDev ArenaChat preference1234EloVerifiedNov 19, 2025Details
Search ArenaChat preference1175EloVerifiedNov 19, 2025Details
EQ-Bench Creative Writing v3Chat preference1317EloVerifiedNov 19, 2025Details
Design ArenaChat preference1142EloVerifiedNov 19, 2025Details
MCP-UniverseTool use40.69%Overall Success RateVerifiedNov 19, 2025Details
MathArena HMMT February 2026Reasoning86.36%accuracyOfficialNov 19, 2025Details
Vectara Hallucination LeaderboardOther19.2%Hallucination RateOfficialNov 19, 2025Details
Vibe Code BenchCoding1.20%Overall accuracyVerifiedNov 19, 2025Details