Vibe Code Bench
A benchmark for measuring the overall capabilities of large language models in code generation and code understanding.
Core Facts
Timeline (last 90 days)
On Vibe Code Bench, Argon scored 91.9 while Opus and Fable scored 90.3, so Argon's advantage is slight and covers only one type of coding task.
In the Vibe Code Bench test of the same app, Argon cost $8.11 and took 37 minutes per task; Sonnet cost $31.25 and took over an hour.
Among the 216 models on Vibe Code Bench, Argon ranks 8th.
Gemini 4 Argon scored 91.9% on the web app development Vibe Code Bench.
In VALS's Vibe Code Bench coding test, Claude Sonnet 5.5 narrowly beat Argon with a score of 92.4 versus 91.9.
Gemini 4 Argon scored 91% on LV Bench (video/long-content understanding), setting a new record.
All Facts (5)
Gemini 4 Argon scored 91% on LV Bench (video/long-content understanding), setting a new record.
70%UnverifiedOn Vibe Code Bench, Argon scored 91.9 while Opus and Fable scored 90.3, so Argon's advantage is slight and covers only one type of coding task.
50%UnverifiedIn the Vibe Code Bench test of the same app, Argon cost $8.11 and took 37 minutes per task; Sonnet cost $31.25 and took over an hour.
50%UnverifiedAmong the 216 models on Vibe Code Bench, Argon ranks 8th.
50%UnverifiedGemini 4 Argon scored 91.9% on the web app development Vibe Code Bench.
50%