dots3-note Preview vs Qwen3.8-27B

Change models
dots3-note PreviewDots Studio
IndexAll categories, out of 100.
63.4 ±3.9 best
59.2 ±2.1
GapIndex points below the top.
Top
−4.2 · no clear gap
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
—
113/s
EloLMArena rating and rank.
—
1436
ReleasedMonth of release.
2026-08
2026-08
Input / 1MUS dollars per 1M input tokens.
Free best
$0.15
Output / 1MUS dollars per 1M output tokens.
Free best
$1.88
1K chats2K tokens in, 500 out, each.
Free best
$1.24
1K documents20K tokens in, 1K out, each.
Free best
$4.88
1K agent tasks200K tokens in, 20K out, each.
Free best
$67.50
ContextMaximum tokens in one request.
512K
1.05M best
AgenticMulti-step tasks with tools.
63.3 best
57.4
CodingCode writing and repair.
60.3
60.3
ReasoningLogic problems and puzzles.
74.5 best
61.2
MultimodalTasks with images and text.
59.2
63.6 best
KnowledgeFacts and expert knowledge.
77.2 best
51.2
MultilingualTasks in many languages.
—
60.6
InstructionTasks with strict prompt rules.
58.2
68.5 best
MathMath problems.
62.3 best
56.3

Benchmarks

10 shared
Only dots3-note Preview: 17Only Qwen3.8-27B: 90
dots3-note Preview
Qwen3.8-27B
76.8% best22 / 131
42.4%52 / 131
75.1% best44 / 86
73.0%47 / 86
61.0%31 / 83
61.7% best27 / 83
52.6% best18 / 77
30.8%45 / 77
91.5% best7 / 59
90.3%8 / 59
80.4% best11 / 57
79.5%15 / 57
NL2RepoCoding
49.8% best12 / 36
42.3%24 / 36
MathVisionMultimodal
87.7%12 / 24
90.0% best9 / 24
83.1%11 / 17
83.7% best9 / 17
BabyVisionMultimodal
50.0%9 / 11
65.7% best6 / 11