DeepSeek V4 Pro 0813 vs dots3-note Preview

Change models
dots3-note PreviewDots Studio
IndexAll categories, out of 100.
65.5 ±3.7 best
63.4 ±3.9
GapIndex points below the top.
Top
−2.1 · no clear gap
CoverageIndex weight that has results.
85%
95% best
SpeedOutput tokens per second.
42/s
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-08
2026-08
Input / 1MUS dollars per 1M input tokens.
$0.111
Free best
Output / 1MUS dollars per 1M output tokens.
$4.44
Free best
Batch output / 1MBatch price per 1M output tokens.
$1.98
—
1K chats2K tokens in, 500 out, each.
$2.44
Free best
1K documents20K tokens in, 1K out, each.
$6.66
Free best
1K agent tasks200K tokens in, 20K out, each.
$111.00
Free best
ContextMaximum tokens in one request.
1.05M best
512K
AgenticMulti-step tasks with tools.
63.7 best
63.3
CodingCode writing and repair.
65.1 best
60.3
ReasoningLogic problems and puzzles.
68.5
74.5 best
MultimodalTasks with images and text.
—
59.2
KnowledgeFacts and expert knowledge.
65.0
77.2 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
62.0 best
58.2
MathMath problems.
64.1 best
62.3

Benchmarks

10 shared
Only DeepSeek V4 Pro 0813: 77Only dots3-note Preview: 17
DeepSeek V4 Pro 0813
dots3-note Preview
61.3%38 / 131
76.8% best22 / 131
80.6% best16 / 114
78.4%27 / 114
87.9% best12 / 86
75.1%44 / 86
55.4%54 / 83
61.0% best31 / 83
42.7%30 / 77
52.6% best18 / 77
BrowseCompAgentic
83.4% best23 / 63
83.3%24 / 63
60.0% best9 / 36
52.6%23 / 36
NL2RepoCoding
61.5% best3 / 36
49.8%12 / 36
74.1% best8 / 23
55.6%19 / 23
89.8%3 / 10
90.9% best1 / 10