DeepSeek V3.1 vs GPT-4.1
Change modelsDeepSeek V3.1DeepSeek
GPT-4.1OpenAI
IndexAll categories, out of 100.
37.9 ±5.8
38.0 ±3.0 best
GapIndex points below the top.
−0.1 · no clear gap
Top
CoverageIndex weight that has results.
85%
100% best
SpeedOutput tokens per second.
98/s best
74/s
EloLMArena rating and rank.
1420 best
1383
ReleasedMonth of release.
2025-08
2025-04
Input / 1MUS dollars per 1M input tokens.
—
$2.00
Output / 1MUS dollars per 1M output tokens.
—
$8.00
Batch output / 1MBatch price per 1M output tokens.
—
$4.00
1K chats2K tokens in, 500 out, each.
—
$8.00
1K documents20K tokens in, 1K out, each.
—
$48.00
1K agent tasks200K tokens in, 20K out, each.
—
$560.00
ContextMaximum tokens in one request.
128K
1.05M best
AgenticMulti-step tasks with tools.
38.0 best
37.8
CodingCode writing and repair.
43.0 best
34.2
ReasoningLogic problems and puzzles.
38.2
40.4 best
MultimodalTasks with images and text.
—
42.1
KnowledgeFacts and expert knowledge.
41.5 best
41.2
MultilingualTasks in many languages.
—
45.1
InstructionTasks with strict prompt rules.
24.3
34.2 best
MathMath problems.
29.9
35.9 best
Benchmarks
13 sharedOnly DeepSeek V3.1: 4Only GPT-4.1: 46
DeepSeek V3.1
GPT-4.1