Skip to content
Leaderboard / Z.ai

GLM-5.1

Z.ai OPEN MIT
Verdict

GLM-5.1 is a solid mid-pack model: #75 of 406 by SIscope score. Strongest at chat quality, weaker at coding. Mid-range on price. Weights are open, so you can run it yourself.

SIscope score
72
#75 of 406
Human-vote rank
#40
of 419 · score 1463
Price per 1M tokens, blended
$2.15
$1.4 in · $4.4 out
Context window
205K
204,800 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsno data yetReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding67
Agents & toolsNo data yet–
Reasoning & knowledge70
Math70
Chat quality91
Following instructions88

What goes into each category · Compare with Gemini 4 Carbon

Human-vote score over time

Text arena, recent updates.

145514631470Oct 20: 1462.7 (rank #39)Oct 22: 1462.8 (rank #39)Oct 28: 1462.8 (rank #40)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

150015081515Oct 16: 1508.5 (rank #57)Oct 18: 1508.1 (rank #58)Oct 19: 1508.1 (rank #60)Oct 22: 1508 (rank #60)Oct 23: 1508 (rank #61)Oct 27: 1507.7 (rank #62)Oct 28: 1507.9 (rank #63)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond90%
PhD-level science multiple choice questions
Better than 84% of 333 tested models. Run Aug 10, 2026.
SimpleQA Verified34%
Short factual questions; rewards not making things up
Better than 37% of 100 tested models. Run Aug 27, 2026.
OTIS Mock AIME 2024-2593%
Competition maths problems in the style of AIME
Better than 77% of 313 tested models. Run Aug 10, 2026.
FrontierMath Tiers 1-337%
Very hard, unpublished research-style maths problems
Better than 45% of 122 tested models. Run Aug 29, 2026.
SWE-bench Verified74%
Fixing real GitHub issues in Python repositories
Better than 53% of 35 tested models. Run May 15, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#40 of 41914631459–146666,771Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#63 of 15215081501–151510,647Oct 28, 2026
Vision arena
Questions about images, blind human votes
Not on this board yet
Agent arena
Real multi-step agent sessions
Not on this board yet

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$1.4
per 1M tokens
Output
$4.4
per 1M tokens
Blended
$2.15
3 in : 1 out, per 1M
Context window
205K
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (z-ai/glm-5.1), Nov 4, 2026. Providers can charge differently.