GLM-5-Turbo vs GLM-5V-Turbo

Change models
IndexAll categories, out of 100.
51.9 ±10.2 best
48.1 ±8.6
GapIndex points below the top.
Top
−3.8 · no clear gap
CoverageIndex weight that has results.
55%
65% best
SpeedOutput tokens per second.
25/s
30/s best
EloLMArena rating and rank.
—
1438
ReleasedMonth of release.
2026-03
2026-04
Input / 1MUS dollars per 1M input tokens.
$1.20
$1.20
Output / 1MUS dollars per 1M output tokens.
$4.00
$4.00
1K chats2K tokens in, 500 out, each.
$4.40
$4.40
1K documents20K tokens in, 1K out, each.
$28.00
$28.00
1K agent tasks200K tokens in, 20K out, each.
$320.00
$320.00
ContextMaximum tokens in one request.
203K
203K
AgenticMulti-step tasks with tools.
57.0 best
50.5
CodingCode writing and repair.
—
—
ReasoningLogic problems and puzzles.
47.0 best
46.8
MultimodalTasks with images and text.
—
49.9
KnowledgeFacts and expert knowledge.
51.2 best
46.9
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
57.4 best
46.1
MathMath problems.
—
—

Benchmarks

8 shared
Only GLM-5V-Turbo: 2
GLM-5-Turbo
GLM-5V-Turbo
27.8% best92 / 222
17.1%121 / 222
71.7% best106 / 217
70.3%112 / 217
0.3%137 / 214
0.6% best132 / 214
84.7% best78 / 199
80.9%97 / 199
28.4%83 / 164
29.3% best80 / 164
98.5%3 / 149
98.5%3 / 149
73.2% best31 / 146
61.1%62 / 146
Claw-EvalAgentic
55.8% best35 / 42
53.8%36 / 42