Quality
#24/6732
Overall quality score
🇨🇳DeepSeekOpen sourceDeepSeek: DeepSeek V3.2.
See how DeepSeek: DeepSeek V3.2 compares with other broadly useful AI models across quality, speed, price, context, and practical capabilities.
Speed
#35/6768 t/s
Output tokens per second
Input price
#30/67$0.27/1M
Per 1M tokens
Output price
#18/67$0.40/1M
Per 1M tokens
Context
#55/67163.8K
Content read at once
Comparison summary
DeepSeek: DeepSeek V3.2 ranks #24 of 67 for overall quality and #35 for response speed in the current catalogue.
Its listed input price is $0.27/1M, output price is $0.40/1M, and it can work with up to 163.8K of context at once.
Its strongest areas in the current data are coding, everyday help, math and logic.
Practical specifications
- Company
- DeepSeek
- Reasoning
- Yes
- Open weights
- Yes
- Input
- Text
- Output
- Text
- Context window
- 163.8K
DeepSeek: DeepSeek V3.2 technical benchmarks.
Named evaluation records for technical comparison. Results retain their original benchmark names and sources.
| Benchmark | Area | DeepSeek V3.2 | Source |
|---|---|---|---|
| Intelligence Index | overallQuality | 32 | OpenRouter / artificial-analysis |
| Coding Index | coding | 44.2 | OpenRouter / artificial-analysis |
| Agentic Index | agentic | 18.3 | OpenRouter / artificial-analysis |
| GPQA | GPQA Diamond | 84 | OpenRouter model benchmarks |
| Humanity's Last Exam | HLE | 22.2 | OpenRouter model benchmarks |
| IFBench | IFBench | 60.7 | OpenRouter model benchmarks |
| τ²-Bench Telecom | τ²-Bench Telecom | 90.6 | OpenRouter model benchmarks |
| AA-LCR | AA-LCR | 65 | OpenRouter model benchmarks |
| GDPval-AA | GDPval-AA | 18.5 | OpenRouter model benchmarks |
| CritPt | CritPt | 2.9 | OpenRouter model benchmarks |
| SciCode | SciCode | 38.9 | OpenRouter model benchmarks |
| Terminal-Bench Hard | Terminal-Bench Hard | 35.6 | OpenRouter model benchmarks |
| AA-Omniscience Accuracy | AA-Omniscience Accuracy | 33.5 | OpenRouter model benchmarks |
| AA-Omniscience Non-Hallucination Rate | AA-Omniscience Non-Hallucination Rate | 18.3 | OpenRouter model benchmarks |
| MMLU-Pro | Knowledge and reasoning | 85 | OpenEvals/leaderboard-data |
| SWE-bench Pro | Coding | 15.56 | OpenEvals/leaderboard-data |
| SWE-bench Verified | Coding | 70 | OpenEvals/leaderboard-data |
| Terminal-Bench | Agentic coding | 39.6 | OpenEvals/leaderboard-data |
| AIME 2026 | Math | 94.17 | OpenEvals/leaderboard-data |
Quality benchmarks
See where DeepSeek: DeepSeek V3.2 stands.
The darker bar marks DeepSeek: DeepSeek V3.2; the lighter bars provide context from other leading models.
Highlights
Quality
Overall capability · Higher is better
Speed
Output tokens per second · Higher is better
Input price
USD per 1M tokens · Lower is better