Comparing 2 models

Gemini 3.1 Pro Preview vs GLM 5.2

Benchmarks through

Verdict

Gemini 3.1 Pro Preview and GLM 5.2 are not yet ranked on the same task, so compare the specifications below. GLM 5.2 costs 285.7x less per input token. Only GLM 5.2 has open weights for self-hosting.

  • Choose GLM 5.2 for you need to self-host.

Built from the published task rankings below; ranks come from each task's own weighted benchmarks, so a win on one task says nothing about another.

metric
Model 1Gemini 3.1 Pro Preview
Model 2GLM 5.2
Best for
Business Writing (#3)
Best Cheap LLM for High Volume (#1)
Provider
google
z-ai
Context
1.0MBest
1.0MBest
Input / 1M
$2
$0.007Best
Output / 1M
$12
$6.80Best
Tool use
yesBest
yesBest
Vision
yesBest
—
Reasoning
yesBest
yesBest
Open-weight
—
yesBest
Released
Feb 19, 2026
Jun 13, 2026

Measured head to head

Benchmarks both models were run on

EvaluationGemini 3.1 Pro PreviewGLM 5.2Source
Design Arena Coding1220 Elo1276 Elosource · 2026-11-15
Design Arena Full-stack Agent1041 Elo1215 Elosource · 2026-11-15
Design Arena UI1215 Elo1286 Elosource · 2026-11-15
Design Arena Web-app Agent1103 Elo1200 Elosource · 2026-11-15
Design Arena Website1225 Elo1277 Elosource · 2026-11-15
FrontierMath Tiers 1-3 (Epoch AI)59.6% ± 2.959.2% ± 3source · 2026-06-11
GPQA Diamond (Epoch AI)94.4% ± 1.691.9% ± 1.6source · 2026-08-06
GPQA Diamond (Epoch AI)94.1% ± 1.791.9% ± 1.6source · 2026-02-20
LiveBench Agentic Coding44.1%51.8%source · 2026-10-18
LiveBench Coding76.5%79.7%source · 2026-10-18
LiveBench Data Analysis78.5%73.7%source · 2026-10-18
LiveBench Instruction Following79.1%62.3%source · 2026-10-18
LiveBench Language85.4%76.2%source · 2026-10-18
LiveBench Math91.0%89.8%source · 2026-10-18
LiveBench Reasoning84%78.6%source · 2026-10-18
LMArena Agent-4.50.2source · 2026-11-11
LMArena Agent Bash Recovery-23.30.7source · 2026-10-08
LMArena Agent Steerability-8.2-1source · 2026-11-11
LMArena Agent Task Outcome-11.31.7source · 2026-11-11
LMArena Agent Tool Hallucination-0.41source · 2026-11-11
LMArena Creative Writing1483 Elo1459 Elosource · 2026-11-11
LMArena Instruction Following1465 Elo1467 Elosource · 2026-11-11
LMArena Long Query1482 Elo1479 Elosource · 2026-11-11
LMArena Math1485 Elo1483 Elosource · 2026-11-11
LMArena Text1487 Elo1475 Elosource · 2026-11-11
LMArena Text Factuality1472 Elo1462 Elosource · 2026-11-11
LMArena WebDev1446 Elo1608 Elosource · 2026-11-11
OTIS Mock AIME (Epoch AI)95.6% ± 3.186.4% ± 4.2source · 2026-08-06
OTIS Mock AIME (Epoch AI)95.6% ± 3.186.4% ± 4.2source · 2026-02-20
SimpleQA Verified (Epoch AI)73.5% ± 1.434.2% ± 1.5source · 2026-08-10
tau2-bench (1.0.1)26%37.1%source · 2026-05-05

Values are as published by each evaluator, in its own units; rows from different evaluators are not on a common scale.