DeepSeek V3.2 (Reasoning)
DeepSeekDeepSeek
Release Date
2025-12-01
Parameters
—
Context Length
131K
Modalities
text
Capability Radar
52
general
54
coding
91
reasoning
56
scienceest.
0
agents
0
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 49 | 73.0 | AA |
| General Ranking | 53 | 78.0 | AA |
| Math Reasoning | 26 | 93.0 | AA |
| Science | 69 | 67.0 | AA |
Benchmark Scores (LLM Stats)
No benchmark data available
AA Evaluation Indices
Math Index92.0
Intelligence Index41.7
Coding Index36.7
Aime 250.9
Tau20.9
Mmlu Pro0.9
Livecodebench0.9
Gpqa0.8
Lcr0.7
Ifbench0.6
Scicode0.4
Terminalbench Hard0.4
Hle0.2
LLM Stats Category Scores
No category score data available
Pricing
Input Price$0.3 / 1M tokens
Output Price$0.45 / 1M tokens
Blended Price (3:1)$0.337 / 1M tokens
Speed
Tokens/sec0.0 tokens/s
Time to First Token0.00s
Time to Answer0.00s
Available Providers
(LS internal units)No provider data available
External Sources
No external links available