DeepSeek-V2.5
Zero-eval
#1DS-FIM-Eval
#1Aider
#1DS-Arena-Code
+5 more
by DeepSeek
+
+
+
+
About
DeepSeek-V2.5 was developed as an enhanced iteration of the DeepSeek-V2 architecture, designed to incorporate improvements in model quality and efficiency. Built to advance the DeepSeek foundation model series, it provides refined capabilities for general-purpose language understanding and generation tasks.
+
+
+
+
Pricing Range
Input (per 1M)$0.14 -$2.00
Output (per 1M)$0.28 -$2.00
Providers3
+
+
+
+
Timeline
AnnouncedMay 8, 2024
ReleasedMay 8, 2024
+
+
+
+
License & Family
License
deepseek
Performance Overview
Performance metrics and category breakdown
Overall Performance
15 benchmarks
Average Score
71.1%
Best Score
95.1%
High Performers (80%+)
6Performance Metrics
Max Context Window
16.4KAvg Throughput
87.7 tok/sAvg Latency
1ms+
+
+
+
All Benchmark Results for DeepSeek-V2.5
Complete list of benchmark scores with detailed information
| GSM8k | text | 0.95 | 95.1% | Self-reported | |
| MT-Bench | text | 0.90 | 90.2% | Self-reported | |
| HumanEval | text | 0.89 | 89.0% | Self-reported | |
| BBH | text | 0.84 | 84.3% | Self-reported | |
| MMLU | text | 0.80 | 80.4% | Self-reported | |
| AlignBench | text | 0.80 | 80.4% | Self-reported | |
| DS-FIM-Eval | text | 0.78 | 78.3% | Self-reported | |
| Arena Hard | text | 0.76 | 76.2% | Self-reported | |
| MATH | text | 0.75 | 74.7% | Self-reported | |
| HumanEval-Mul | text | 0.74 | 73.8% | Self-reported |
Showing 1 to 10 of 15 benchmarks
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+