Skip to main content

DeepSeek V3.2 Speciale

DeepSeekDeepSeekOpen WeightMIT · Commercial OK

Description

DeepSeek-V3.2-Speciale is a specialized variant of DeepSeek-V3.2, optimized for enhanced performance on specific tasks.

Release Date
2025-12-01
Parameters
685.0B
Context Length
Modalities
text

Capability Radar

43
general
79
coding
95
reasoning
60
science
50
agents
0
multimodal

Rankings

Domain#RankScoreSource
Agentic Capability124
34.0
LS
Code Ranking72
76.0
AA
General Ranking249
48.0
AA
Science79
72.0
AA

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Agents

t2-bench80.3%SR
Terminal-Bench 2.0Stanford × Laude Institute (2026)46.4%SR
Toolathlon35.2%SR

Code

SWE-Bench Verified73.1%SR

Math

HMMT 202599.2%SR
AIME 202596.0%SR
CodeForces0.90 / 3000SR
Humanity's Last Exam30.6%SR

AA Evaluation Indices

(Artificial Analysis)
Math Index(Artificial Analysis)
96.7
Intelligence Index(Artificial Analysis)
22.6
Aime 25(MAA (Mathematical Association of America))
1.0
Livecodebench(UC Berkeley + MIT + Cornell (2024))
0.9
Gpqa(NYU + Cohere + Anthropic (2023))
0.9
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))
0.9
Lcr(Artificial Analysis)
0.6
Ifbench(Google Research (2023))
0.6
Scicode(UIUC + Argonne National Lab (2024))
0.4
Terminalbench Hard(Stanford × Laude Institute (2026))
0.3
Hle(Center for AI Safety + Scale AI (2025))
0.3
Tau2(Sierra + U Toronto + Vector Institute (2025))
0.0

LLM Stats Category Scores

(LLM Stats (zeroeval))
Math
80
Frontend Development
70
General
70
Reasoning
60
Code
60
Agents
50
Tool Calling
50
Vision
30

Pricing

Input PriceFree
Output PriceFree
Blended Price (3:1)Free

Speed

Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s

Provider Price Ranking

No provider data available

External Sources