DeepSeek V4 Pro vs Mistral Large 4

Change models
IndexAll categories, out of 100.
60.3 ±3.8 best
60.1 ±4.0
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
85%
85%
SpeedOutput tokens per second.
96/s best
57/s
EloLMArena rating and rank.
1451 best
1413
ReleasedMonth of release.
2026-08
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.90
$0.70 best
Output / 1MUS dollars per 1M output tokens.
$1.50 best
$2.10
1K chats2K tokens in, 500 out, each.
$2.55
$2.45 best
1K documents20K tokens in, 1K out, each.
$19.50
$16.10 best
1K agent tasks200K tokens in, 20K out, each.
$210.00
$182.00 best
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
61.4
65.5 best
CodingCode writing and repair.
60.0
62.1 best
ReasoningLogic problems and puzzles.
58.6 best
58.3
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
62.4 best
44.1
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
56.0
57.2 best
MathMath problems.
56.7
60.7 best

Benchmarks

29 shared
Only DeepSeek V4 Pro: 54Only Mistral Large 4: 32
DeepSeek V4 Pro
Mistral Large 4
89.6% best48 / 220
89.1%54 / 220
18.0% best34 / 207
10.0%64 / 207
96.7%39 / 203
98.9% best25 / 203
49.1% best26 / 170
26.0%100 / 170
54.5% best23 / 132
46.1%37 / 132
49.9%73 / 130
78.4% best40 / 130
MedScribeHealthcare
75.1%83 / 111
80.4% best59 / 111
MedCodeHealthcare
40.5%65 / 108
40.7% best62 / 108
23.1%73 / 98
31.7% best58 / 98
47.0% best40 / 94
21.2%80 / 94
Chess PuzzlesReasoning
19.0%71 / 93
31.0% best53 / 93
45.3%61 / 93
58.6% best48 / 93
25.2%47 / 89
30.4% best36 / 89
17.0%54 / 83
18.5% best51 / 83
74.5%41 / 80
76.5% best33 / 80
82.7%51 / 80
83.9% best43 / 80
44.1%58 / 79
54.7% best26 / 79
26.2%46 / 77
30.6% best41 / 77
EMBFinance
51.6%55 / 75
56.2% best47 / 75
2.4%68 / 75
26.8% best42 / 75