readable AI benchmarks (simplified)
DeepSeek
DeepSeek V3.2 (Reasoning)
DeepSeek V3.2 (Reasoning) has a Quality Score of 30.1, ranking 156th among 238 scored models. Per 1M tokens, pricing is $0.280 input, $0.420 output, and $0.014 cached input. It has a 128k token context window. Its Reliability Score is 38.8, ranking 172nd among 245 scored models, below average (total average: 46.4). Its Value Score is 26.9, around average compared with the total Value average of 28.3.
Quality Score30.1coverage 100.0%: AA Intelligence and AA-Omniscience outcomes available
Value Score26.9
Factual reliability38.8
Cache Discount95%
Model specification
- Reasoning
- reasoning
- Input modalities
- none
- Output modalities
- none
- Context window
- 128k tokens
- Weights
- open weights
Token prices USD per 1M tokens
- Input
- $0.280
- Output
- $0.420
- Cached Input
- $0.014
- Cache write
- not available
- Cost per task
- not available
Capability
- Intelligence
- 21.5
- Omniscience (native net)
- -22.5
- Correct
- 33.0%
- Blended price
- $0.108
Answer outcomes
Fully graded outcomes (Correct + Incorrect): 88.4%
Correct 33.0%Incorrect 55.5%Partial / not attempted 11.6%
Artificial Analysis benchmarks legacy records
- Legacy GDPval-AA (normalized)
- 10%
- Humanity’s Last Exam
- 25%
- GPQA Diamond
- 84%
- CritPt
- 3%
- AA-LCR
- 73%
Similar models
- DeepSeek V4 Flash 0420 (High)DeepSeekQuality Score 31.4Value Score 29.1
- DeepSeek V4 Pro 0813 (Max)DeepSeekQuality Score 34.9Value Score 25.6
- DeepSeek V3.2 Exp (Reasoning)DeepSeekQuality Score 25.5Value Score 22.6
- DeepSeek V3.1 Terminus (Reasoning)DeepSeekQuality Score 25.8Value Score 16.9
- DeepSeek V3.1 (Reasoning)DeepSeekQuality Score 24.3Value Score 17.8
- DeepSeek V4 Pro 0424 (High)DeepSeekQuality Score 37.4Value Score 31.9