Claude Opus 4.5 vs dots3-note Preview

Change models
dots3-note PreviewDots Studio
IndexAll categories, out of 100.
52.0 ±2.1
63.4 ±3.9 best
GapIndex points below the top.
−11.4
Top
CoverageIndex weight that has results.
100% best
95%
SpeedOutput tokens per second.
39/s
—
EloLMArena rating and rank.
1451
—
ReleasedMonth of release.
2025-11
2026-08
Input / 1MUS dollars per 1M input tokens.
$5.00
Free best
Output / 1MUS dollars per 1M output tokens.
$25.00
Free best
Batch output / 1MBatch price per 1M output tokens.
$12.50
—
1K chats2K tokens in, 500 out, each.
$22.50
Free best
1K documents20K tokens in, 1K out, each.
$125.00
Free best
1K agent tasks200K tokens in, 20K out, each.
$1,500.00
Free best
ContextMaximum tokens in one request.
200K
512K best
AgenticMulti-step tasks with tools.
54.7
63.3 best
CodingCode writing and repair.
55.0
60.3 best
ReasoningLogic problems and puzzles.
51.8
74.5 best
MultimodalTasks with images and text.
42.2
59.2 best
KnowledgeFacts and expert knowledge.
54.6
77.2 best
MultilingualTasks in many languages.
56.6
—
InstructionTasks with strict prompt rules.
41.2
58.2 best
MathMath problems.
51.3
62.3 best

Benchmarks

13 shared
Only Claude Opus 4.5: 88Only dots3-note Preview: 14
Claude Opus 4.5
dots3-note Preview
80.9% best13 / 114
78.4%27 / 114
57.1%45 / 83
61.0% best31 / 83
30.8%45 / 77
52.6% best18 / 77
90.9%24 / 70
93.9% best10 / 70
84.8%22 / 59
91.5% best7 / 59
58.0%45 / 57
80.4% best11 / 57
70.6%44 / 53
79.1% best18 / 53
Claw-EvalAgentic
59.6%28 / 42
73.4% best6 / 42
NL2RepoCoding
43.2%21 / 36
49.8% best12 / 36
MathVisionMultimodal
74.3%23 / 24
87.7% best12 / 24
SimpleVQAMultimodal
65.7%11 / 18
72.5% best6 / 18
76.4%8 / 16
78.9% best7 / 16
VideoMMMUMultimodal
84.4%
86.8%