DeepSeek
DeepSeek R1 Distill Llama 70B
DeepSeek R1 Distill Llama 70B is a distilled large language model based on [Llama-3.3-70B-Instruct](/meta-llama/llama-3.3-70b-instruct), using outputs from [DeepSeek R1](/deepseek/deepseek-r1). The model combines advanced distillation techniques to achieve high performance across...
DeepSeek release notesModel Snapshot
ReleaseJan 20, 2025
Context8,192
Overall
Rank #362 across 566
9.9
Coding
No rank
n/a
Blended Price
Rank #152 across 371
$0.787/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed22.6 tok/s
First Token0.62s
Max Output8,192
Pricing
Blended$0.787/M
Input$0.700/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 9.9 | #362 |
| Artificial Analysis Math Index | math | 53.7 | #133 |
| MMLU-Pro | reasoning | 79.5% | #128 |
| GPQA | reasoning | 40.2% | #450 |
| Humanity's Last Exam |