Rankings / xAI
Grok Code Fast 1
released 2025-08-28
BenchAtlas Index
as of 2026-09-11
base configuration
rank #294
9 families · 5 categories · high
Benchmark evidence
11 results
agentic coding
| τ²-Bench | 75.7 % | independent | Artificial Analysis |
coding
| LiveCodeBench v6 implementation=artificial-analysis | 65.7 % | independent | Artificial Analysis | |
| SciCode | 36.2 % | independent | Artificial Analysis |
external indices
| Intelligence Index v4.1 | 14.1 points | independent | Artificial Analysis | |
| AA Math Index | 43.3 points | independent | Artificial Analysis |
knowledge science
| GPQA Diamond implementation=artificial-analysis | 72.7 % | independent | Artificial Analysis | |
| Humanity's Last Exam implementation=artificial-analysis | 8.0 % | independent | Artificial Analysis | |
| MMLU-Pro implementation=artificial-analysis | 79.3 % | independent | Artificial Analysis |
long context instruction
| AA-LCR | 53.0 % | independent | Artificial Analysis | |
| IFBench | 41.4 % | independent | Artificial Analysis |
reasoning math
| AIME year=2025 · implementation=artificial-analysis | 43.3 % | independent | Artificial Analysis |
Agent + model results
systems, not bare-model scores
| agent + model mini-SWE-agent + Grok Code Fast 1 | Terminal-Bench 2.0 | 25.8 % | community | Terminal-Bench Leaderboard |
| agent + model Artificial Analysis harness + Grok Code Fast 1 | Terminal-Bench Hard | 17.4 % | independent | Artificial Analysis |
| agent + model terminus-2 + Grok Code Fast 1 | Terminal-Bench 2.0 | 14.2 % | community | Terminal-Bench Leaderboard |
These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.