Models/gemini-3-1-pro

Gemini 3.1 Pro Preview

18th on Convex No Guidelines at 72.32%. With Guidelines is 84.82%; that row has no No Guidelines + Web run. Its best ReactBench v1.2 row is high at 25.6%.

Google · Proprietary · Fetched 2026-10-10

LMArena WebDev

1447

#81 / 153 · ±5

LMArena Image-to-WebDev

1481

#46 / 65 · ±9

The Slop Index

#6 / 58

Benchmark scores

Boards older than 90 days are not counted as recommendations

BenchmarkSettingScoreRankCostAs of

LMArena WebDev—1447±5#81—2026-10-27

LMArena Image-to-WebDev—1481±9#46—2026-10-19

The Slop Index—71#6—2026-10-28

EQ-Bench Creative Writing v3—1491.2#69—2026-10-29

Convex Evals1 run72.32#18$3.992026-10-28

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 71

Peers

ModelWebDevPrice $/M

Gemini 3.1 Pro PreviewCurrent1447—GPT 5.51512—DeepSeek V4 Pro1582—GPT 5.41463—