dots3-note Preview vs Grok 4.20

Change models
dots3-note PreviewDots Studio
IndexAll categories, out of 100.
63.4 ±3.9 best
52.8 ±4.8
GapIndex points below the top.
Top
−10.6
CoverageIndex weight that has results.
95% best
90%
SpeedOutput tokens per second.
—
72/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-08
2026-03
Input / 1MUS dollars per 1M input tokens.
Free best
$1.25
Output / 1MUS dollars per 1M output tokens.
Free best
$2.50
1K chats2K tokens in, 500 out, each.
Free best
$3.75
1K documents20K tokens in, 1K out, each.
Free best
$27.50
1K agent tasks200K tokens in, 20K out, each.
Free best
$300.00
ContextMaximum tokens in one request.
512K
2M best
AgenticMulti-step tasks with tools.
63.3 best
50.7
CodingCode writing and repair.
60.3 best
55.4
ReasoningLogic problems and puzzles.
74.5 best
56.0
MultimodalTasks with images and text.
59.2 best
45.8
KnowledgeFacts and expert knowledge.
77.2 best
55.4
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
58.2
—
MathMath problems.
62.3 best
57.4

Benchmarks

5 shared
Only dots3-note Preview: 22Only Grok 4.20: 17
dots3-note Preview
Grok 4.20
78.4% best27 / 114
76.7%40 / 114
61.0% best31 / 83
51.8%65 / 83
79.1% best18 / 53
75.2%36 / 53
92.1% best8 / 24
62.8%23 / 24
SimpleVQAMultimodal
72.5% best6 / 18
57.4%14 / 18