Reddit users report Gemini Argon underperforming Gemini 3.1 Pro on instructions and audio
PreparationOdd430 · reddit · 2026-11-02
A Reddit post details a perceived regression after upgrading to Gemini Argon: confidently wrong answers, following only 60% of clear instructions, and failing basic audio transcription — a screenshot shows the model repeatedly claiming uploaded MP3 attachments came through empty. The author says Gemini 3.1 Pro handled the same prompts fine and asks for workarounds before switching back. Unverified user-side observation, no official response.
More from Models
- Mistral employee shows its model autonomously picked STO-3G basis set, hit 1e-13 accuracy vs PySCF — MistralAI · 2026-11-02
- Fable 5.5 launch week teased: "see you real soon" — iruletheworldmo · 2026-11-02
- Amazon CEO expects just 6-8 AI models to be 'good enough' for most enterprise work — YvesMulkers · 2026-11-02
- ZenteiQ, backed by IndiaAI Mission, to unveil homegrown foundation models tomorrow — deepakns · 2026-11-02
- Altman: model that cracked Navier-Stokes is the next GPT, run as 10,000 agents for 4 days — Hesamation · 2026-11-02
- Tesla FSD Nears 17 Billion Miles, Adding 1.8M Miles Per Hour — XFreeze · 2026-11-02