by DeepSeek
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 multiple benchmarks, including: - AIME 2024 pass@1: 70.0 - MATH-500 pass@1: 94.5 - CodeForces Rating: 1633 The model leverages fine-tuning from DeepSeek R1's outputs, enabling competitive performance comparable to larger frontier models.
| Signal | Strength | Weight | Impact |
|---|---|---|---|
| Context Window2026-03-03T20:28:12.458Z | 81 | 15% | +12.2 |
| Capabilities2026-03-03T20:28:12.458Z | 43 | 25% | +10.7 |
| Recency2026-03-03T20:28:12.458Z | 59 | 15% | +8.9 |
| Output Capacity2026-03-03T20:28:12.458Z | 70 | 10% | +7.0 |
| Versatility2026-03-03T20:28:12.458Z | 33 | 10% | +3.3 |
| Pricing Tier2026-03-03T20:28:12.458Z | 1 | 25% | +0.2 |
Cost Estimator
You save $33.49/month vs category average