Gemini 3.1 Pro: what developers actually say
We read every Hacker News comment from Nov 2025 – Nov 2026 that gave a real opinion on Gemini 3.1 Pro. 31 of the 36 opinions come from people who say they use it themselves.
Developers give Gemini 3.1 Pro a 54% approval rating across 36 opinions from 34 authors, with 31 firsthand accounts. Code quality draws more praise than criticism at 10 positive and 6 negative mentions, while reliability is weak at 1 positive and 5 negative. Monthly approval fell to 40% in July on 5 opinions and rose to 75% in October on 2 opinions. The recent samples are small, so the trend is noisy.
Pick it if
- Want an inexpensive chat plan for pasting code
- Need a second opinion on system design
- Build single-file prototypes or convert images to HTML
Look elsewhere if
- Need dependable tool calling for agent workflows
- Run security reviews and want few false positives
- Need predictable token use during agent runs
Strengths and weaknesses
Each row counts opinions that mentioned the topic. Bars to the right of the line are praise, bars to the left are complaints.
How opinion has moved
What developers say
Praise
Complaints
Summaries are paraphrased. Each links to the original comment.
FAQ
Is Gemini 3.1 Pro worth it?
Approval is 54% from 36 opinions, so reactions are split. Code quality leans positive, but reliability leans negative. Its strongest reported uses are chat-based coding, UI work, and second opinions. Agentic coding and security reviews draw more complaints.
Is Gemini 3.1 Pro better than Claude Opus 4.6?
The evidence is thin and split. One developer preferred Gemini 3.1 Pro over Claude Opus 4.6, and one preferred Claude Opus 4.6 instead. The data does not show a clear winner.
Is Gemini 3.1 Pro good for agentic coding?
The data points to problems here. Reviewers cite weak tool calling, poor task adherence, heavy token use, and one runaway token incident. Reliability mentions include 1 positive and 5 negative.