Skip to content
Leaderboard / Z.ai

GLM-4.7

Z.ai OPEN MIT
Verdict

GLM-4.7 is a solid mid-pack model: #125 of 406 by SIscope score. Strongest at chat quality, weaker at reasoning & knowledge. Mid-range on price. Weights are open, so you can run it yourself.

SIscope score
62
#125 of 406
Human-vote rank
#89
of 419 · score 1435
Price per 1M tokens, blended
$1
$0.6 in · $2.2 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.

Coding60
Agents & toolsNo data yet–
Reasoning & knowledge57
Math67
Chat quality79
Following instructions75

What goes into each category · Compare with Gemini 4 Carbon

Human-vote score over time

Text arena, recent updates.

143014351440Oct 20: 1435.3 (rank #87)Oct 22: 1435.4 (rank #87)Oct 28: 1435.4 (rank #89)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

143014351440Oct 16: 1435.3 (rank #82)Oct 18: 1435.4 (rank #82)Oct 19: 1435.3 (rank #84)Oct 22: 1435.4 (rank #84)Oct 23: 1435.4 (rank #85)Oct 27: 1435.4 (rank #87)Oct 28: 1435.2 (rank #88)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 Diamond83%
PhD-level science multiple choice questions
Better than 64% of 333 tested models. Run Jan 29, 2026.
SimpleQA Verified32%
Short factual questions; rewards not making things up
Better than 29% of 100 tested models. Run Aug 27, 2026.
OTIS Mock AIME 2024-2583%
Competition maths problems in the style of AIME
Better than 63% of 313 tested models. Run Jan 29, 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
#89 of 41914351429–144212,239Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#88 of 15214351423–14473,948Oct 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
$0.6
per 1M tokens
Output
$2.2
per 1M tokens
Blended
$1
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-4.7), Nov 4, 2026. Providers can charge differently.