DeepSeek V4 Flash Vision Exp vs Mistral Large 4

Change models
IndexAll categories, out of 100.
59.6 ±7.2
60.1 ±4.0 best
GapIndex points below the top.
−0.5 · no clear gap
Top
CoverageIndex weight that has results.
85%
85%
SpeedOutput tokens per second.
31/s
57/s best
EloLMArena rating and rank.
—
1413
ReleasedMonth of release.
2026-08
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.216 best
$0.70
Output / 1MUS dollars per 1M output tokens.
$0.647 best
$2.10
Batch output / 1MBatch price per 1M output tokens.
$0.33
—
1K chats2K tokens in, 500 out, each.
$0.755 best
$2.45
1K documents20K tokens in, 1K out, each.
$4.96 best
$16.10
1K agent tasks200K tokens in, 20K out, each.
$56.06 best
$182.00
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
70.4 best
65.5
CodingCode writing and repair.
43.1
62.1 best
ReasoningLogic problems and puzzles.
67.2 best
58.3
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
64.5 best
44.1
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
68.9 best
57.2
MathMath problems.
61.2 best
60.7

Benchmarks

12 shared
Only Mistral Large 4: 49
DeepSeek V4 Flash Vision Exp
Mistral Large 4
79.5% best14 / 80
76.5%33 / 80
85.4% best38 / 80
83.9%43 / 80
65.1% best5 / 66
57.2%22 / 66
68.2%63 / 66
77.2% best37 / 66
71.0% best24 / 66
64.8%42 / 66
80.4% best27 / 66
49.6%65 / 66
87.8%42 / 66
93.6% best18 / 66
37.6%51 / 52
60.0% best47 / 52
70.7% best38 / 52
65.6%43 / 52
69.2% best35 / 52
28.1%52 / 52
39.0% best24 / 51
11.8%48 / 51
64.0% best25 / 49
8.0%45 / 49