GLM-4.5 vs Grok 4 Fast (Reasoning)
Change models
IndexAll categories, out of 100.
45.7 ±9.2 best
45.3 ±4.3
GapIndex points below the top.
Top
−0.4 · no clear gap
CoverageIndex weight that has results.
65%
95% best
SpeedOutput tokens per second.
25/s
—
EloLMArena rating and rank.
1430 best
1400
ReleasedMonth of release.
2025-07
2025-09
Input / 1MUS dollars per 1M input tokens.
$0.60
—
Output / 1MUS dollars per 1M output tokens.
$2.20
—
1K chats2K tokens in, 500 out, each.
$2.30
—
1K documents20K tokens in, 1K out, each.
$14.20
—
1K agent tasks200K tokens in, 20K out, each.
$164.00
—
ContextMaximum tokens in one request.
131K
2M best
AgenticMulti-step tasks with tools.
48.1 best
42.9
CodingCode writing and repair.
43.3
44.9 best
ReasoningLogic problems and puzzles.
—
47.8
MultimodalTasks with images and text.
—
42.7
KnowledgeFacts and expert knowledge.
45.1
46.6 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
36.2
MathMath problems.
48.4
55.4 best
Benchmarks
11 sharedOnly GLM-4.5: 5Only Grok 4 Fast (Reasoning): 24
GLM-4.5
Grok 4 Fast (Reasoning)