DeepSeek V3.2 (Thinking) vs Gemini 3.1 Flash Lite Preview
Change modelsDeepSeek V3.2 (Thinking)DeepSeek
IndexAll categories, out of 100.
49.4 ±8.6
49.6 ±6.9 best
GapIndex points below the top.
−0.2 · no clear gap
Top
CoverageIndex weight that has results.
65%
80% best
SpeedOutput tokens per second.
—
151/s
EloLMArena rating and rank.
1420 best
1416
ReleasedMonth of release.
2025-12
2026-03
Input / 1MUS dollars per 1M input tokens.
$0.55
$0.25 best
Output / 1MUS dollars per 1M output tokens.
$2.19
$1.50 best
1K chats2K tokens in, 500 out, each.
$2.20
$1.25 best
1K documents20K tokens in, 1K out, each.
$13.19
$6.50 best
1K agent tasks200K tokens in, 20K out, each.
$153.80
$80.00 best
ContextMaximum tokens in one request.
128K
1.05M best
AgenticMulti-step tasks with tools.
47.7 best
41.3
CodingCode writing and repair.
49.2 best
48.3
ReasoningLogic problems and puzzles.
—
—
MultimodalTasks with images and text.
—
62.5
KnowledgeFacts and expert knowledge.
51.2
52.3 best
MultilingualTasks in many languages.
—
57.3
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
51.0
51.9 best
Benchmarks
12 sharedOnly DeepSeek V3.2 (Thinking): 5Only Gemini 3.1 Flash Lite Preview: 19
DeepSeek V3.2 (Thinking)
Gemini 3.1 Flash Lite Preview