Side-by-side comparison · models
Grok 3 vs Grok 4.1 Fast Non-Reasoning
What we have checked about each, line by line. Every value shows where it came from and when we last confirmed it.
Models
Grok 3
Grok model for agentic tool use, reasoning, coding, and live assistance.
Fast Grok model for responsive chat, reasoning, and tool-assisted work.
Checked facts
Line by line
| Measure | Grok 3 | Grok 4.1 Fast Non-Reasoning |
|---|---|---|
| Time to first token (ms) | 1,430 | Not checked yet |
| Output speed (tokens a second) | 75.4 | Not checked yet |
| Gpqa diamond | 75.8% (for 9 Apr 2025) | Not checked yet |
| Arc agi 2 | 0% (for 9 Apr 2025) | Not checked yet |
| Aider polyglot | 53.3% (for 9 Apr 2025) | Not checked yet |
| Arc agi | 5.5% (for 9 Apr 2025) | Not checked yet |
| Balrog | 29.5% (for 9 Apr 2025) | Not checked yet |
| Fiction livebench | 58.3% (for 9 Apr 2025) | Not checked yet |
| Frontiermath 2025 02 28 private | 3.8% (for 9 Apr 2025) | Not checked yet |
| Frontiermath tier 4 2025 07 01 private | 0% (for 9 Apr 2025) | Not checked yet |
| Lech mazur writing | 76.4% (for 9 Apr 2025) | Not checked yet |
| Math level 5 | 88.7% (for 9 Apr 2025) | Not checked yet |
| Simplebench | 36.1% (for 9 Apr 2025) | Not checked yet |
| Weirdml | 37.2% (for 9 Apr 2025) | Not checked yet |
| Otis mock aime 2024 2025 | 55.6% (saturated benchmark: the top models all score near the ceiling) (for 9 Apr 2025) | Not checked yet |
A line appears only when at least one record has a value we confirmed from a source. Values we have not confirmed are left out, never guessed. Records of different types share fewer lines.
Add a third
Compare with one more record
How this works
Share it, or build your own
This page works for any two or three records in the directory, of any type. The link stays the same whichever order you name them in, so you can share it as it is. Change the records in the address, for example /directories/vs/claude-code-vs-cursor, or start from our editorial comparisons.