vibecodingnav
中/EN
Models/gemini-3-1-pro

Gemini 3.1 Pro Preview

7th on Convex No Guidelines at 72.32%. That row has no No Guidelines + Web run. Its best ReactBench v1.2 row is high at 25.6%, 54th of 76.

Google · Proprietary · Fetched 2026-10-10

LMArena WebDev
1447
#72 / 142 · ±5
LMArena Image-to-WebDev
1481
#46 / 64 · ±9
The Slop Index
66
#6 / 39

Benchmark scores

Boards older than 90 days are not counted as recommendations
BenchmarkSettingScoreRankCostAs of
LMArena WebDev—1447±5#72—2026-10-08
LMArena Image-to-WebDev—1481±9#46—2026-10-09
The Slop Index—66#6—2026-10-11
Convex Evals1 run72.32#7$3.992026-10-08
ReactBenchhigh25.622.9–28.4#54$0.802026-10-10
DeepSWE v1.1high · mini-swe-agent11.710.2–13.2#28$2.142026-09-22
SWE-bench Verified—This board has stopped updating——2026-02-26
Aider Polyglot—This board has stopped updating——2025-11-20

Use it or not

Why
  • Still mid-board on the Convex baseline, at $3.99.
When not
  • No web-arm score. A missing run is not a zero, and it is not added to the Convex score.

Appears in these scenarios

Collected from scenario boards
Build a Web frontend#28DeepSWE 11.7%Build an Android app#28DeepSWE 11.7%Write CLI, scripts, and DevOps#28DeepSWE 11.7%Write docs & copy#6Slop Index 66