DeepSeek V3.2 vs o1

Change models
o1OpenAI
IndexAll categories, out of 100.
44.0 ±4.9
44.3 ±4.1 best
GapIndex points below the top.
−0.3 · no clear gap
Top
CoverageIndex weight that has results.
85%
95% best
SpeedOutput tokens per second.
36/s
97/s best
EloLMArena rating and rank.
1425 best
1366
ReleasedMonth of release.
2025-12
2024-12
Input / 1MUS dollars per 1M input tokens.
$0.26 best
$15.00
Output / 1MUS dollars per 1M output tokens.
$0.38 best
$60.00
1K chats2K tokens in, 500 out, each.
$0.71 best
$60.00
1K documents20K tokens in, 1K out, each.
$5.58 best
$360.00
1K agent tasks200K tokens in, 20K out, each.
$59.60 best
$4,200.00
ContextMaximum tokens in one request.
164K
200K best
AgenticMulti-step tasks with tools.
43.1
45.0 best
CodingCode writing and repair.
48.4 best
30.9
ReasoningLogic problems and puzzles.
42.7
43.4 best
MultimodalTasks with images and text.
—
55.3
KnowledgeFacts and expert knowledge.
43.3
46.6 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
35.0
51.1 best
MathMath problems.
46.4 best
43.5

Benchmarks

17 shared
Only DeepSeek V3.2: 21Only o1: 12
DeepSeek V3.2
o1
77.3% best99 / 220
76.8%102 / 220
11.2% best132 / 216
7.0%155 / 216
45.7%157 / 211
65.0% best130 / 211
0.9% best119 / 207
0.3%132 / 207
68.4%106 / 203
73.3% best97 / 203
75.1% best112 / 193
74.7%115 / 193
24.0%111 / 170
34.5% best72 / 170
66.0%135 / 149
80.4% best86 / 149
78.9% best68 / 143
62.6%79 / 143
69.9% best91 / 142
50.3%116 / 142
49.0%76 / 140
70.3% best41 / 140
76.3% best78 / 135
73.2%86 / 135
MMLU ProKnowledge
83.1%74 / 135
83.5% best71 / 135
AIMEMath
64.8%55 / 95
71.5% best53 / 95
MedQAHealthcare
89.5%51 / 95
96.5% best1 / 95
MGSMMultilingual
90.9% best31 / 74
89.3%47 / 74
22.1% best27 / 69
9.3%38 / 69