gemini-4POST
Tester questions near-identical GPT-6 Pro and Gemini 4 Pro results across different harnesses
The poster notes that GPT-6 Pro was evaluated in the ChatGPT harness and Gemini 4 Pro in the arena.ai harness, yet the two different models produced basically identical results, and questions the evaluation method.
Raphi-2Code
@R2Cdev_Archived 2026.09.20Originally posted 2026.09.20
Raphi-2Code @R2Cdev_Originally posted 2026.09.20
Why are the results basically identical when they were evaluated in different harnesses (GPT-6 Pro in the ChatGPT harness, Gemini 4 Pro in the https://arena.ai harness) and are completely different models?
This post quotes another post: “Poll: Gemini 4 Pro vs GPT-6 Astra image generation”
x.com/r2cdev_/status/2101409563094233293Archived 2026.09.20 10:14