Intelligence Index
019.8
> 15% OF MODELS
DeepSeek
DeepSeek R1 Distill Llama 70B
Updated: September 2026
Distilled Llama model for efficient, quality output.
Specifications
- Context
- 131K
- Input Modalities
- Text, File
- Output Modalities
- Text
- Input
- $0.7/M
- Output
- $0.8/M
Capabilities
THINKING
Similarly Priced Models
| Model | Provider | Context | Input Price | Output Price |
|---|---|---|---|---|
| R1 | DeepSeek | 64K | $0.7/M | $2.5/M |
| Qwen2.5 Coder 32B Instruct | Qwen | 33K | $0.66/M | $1/M |
| Kimi K2.6 | Moonshot | 256K | $0.7448/M | $4.655/M |
| Gemma 2 27B | Google | 8K | $0.65/M | $0.65/M |
| Gemini 3.8 Flash | Google | 1M | $0.75/M | $3.75/M |
Performance Metrics
Average Response Performance
- Output Speed
- 25.7 tok/s
- Time To First Token
- 1.06s
- Time To First Answer Token
- 78.78s
- End To End Response Time
- 98.21s
DATA SOURCE: Artificial Analysis