DeepSeek V4 Pro 0423 vs Qwen 3.6 Max (preview)

Change models
IndexAll categories, out of 100.
57.3 ±4.8 best
56.9 ±5.0
GapIndex points below the top.
Top
−0.4 · no clear gap
CoverageIndex weight that has results.
85%
85%
SpeedOutput tokens per second.
39/s
75/s best
EloLMArena rating and rank.
1451 best
1446
ReleasedMonth of release.
2026-04
2026-04
Input / 1MUS dollars per 1M input tokens.
$0.90 best
$1.03
Output / 1MUS dollars per 1M output tokens.
$1.50 best
$6.16
1K chats2K tokens in, 500 out, each.
$2.55 best
$5.14
1K documents20K tokens in, 1K out, each.
$19.50 best
$26.70
1K agent tasks200K tokens in, 20K out, each.
$210.00 best
$328.64
ContextMaximum tokens in one request.
1.05M best
262K
AgenticMulti-step tasks with tools.
55.6
59.0 best
CodingCode writing and repair.
58.9 best
56.8
ReasoningLogic problems and puzzles.
54.3 best
53.1
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
59.6 best
59.1
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
52.6
60.4 best
MathMath problems.
57.3 best
53.0

Benchmarks

10 shared
Only DeepSeek V4 Pro 0423: 31Only Qwen 3.6 Max (preview): 18
DeepSeek V4 Pro 0423
Qwen 3.6 Max (preview)
89.6% best45 / 217
87.4%64 / 217
96.7% best36 / 200
91.1%58 / 200
CorpFin v2Finance
61.4%52 / 131
66.5% best18 / 131
47.0%37 / 91
52.0% best27 / 91
SWE-benchCoding
77.4% best35 / 87
72.8%54 / 87
17.0%51 / 80
19.0% best46 / 80
56.2% best14 / 66
51.7%18 / 66
59.4% best27 / 53
47.9%50 / 53
60.4% best4 / 51
52.8%22 / 51
77.6% best5 / 32
76.7%9 / 32