gemini-4IMAGE

Gemini-4-Argon joins RSI-Inference Bench: 1.39× speedup, 6th overall

Gemini-4-Argon enters the RSI-Inference Bench with a 1.39× speedup (4th of 14), 6th overall and 6.7% model-support pass@1, making Google the third provider with a non-zero score.

Junteng Liu

@junteng88716710Archived 1 day agoOriginally posted 2026.10.26

Gemini-4-Argon joins RSI-Inference Bench: 1.39× speedup, 6th overall

Junteng Liu @junteng88716710Originally posted 2026.10.26

Open original on X

🆕 @GeminiApp’s Gemini-4-Argon joins RSI-Inference Bench! ⚡ 1.39× speedup (#4/14) 🏆 #6/14 overall 🛠️ 6.7% model-support pass@1 Google is the third provider, after OpenAI and Anthropic, to achieve non-zero model-support pass@1 on our benchmark. https://rsi-inference.com

This post quotes another post: “Apple releases RSI-Inference Bench for inference engineering”

x.com/junteng88716710/status/2114653623468867710Archived 2026.10.27 09:54