DeepSeek
R1 Distill 14B Qwen
reasoning
14,8B
Parameters
8,4 Go
Q4_K_M file
≥ 9,9 Go
Recommended VRAM
131 072
Max context
Speed by graphics card
Simulate in the “Configurator” →Standard llama-bench run (prompt 512 / generation 128 tokens), model fully in VRAM. Click a card for details. Electricity cost at the default price (0,25 €/kWh).
| Card · Q4_K_M | VRAM | Generation | Prompt | VRAM used | Power | € elec. / M tok | Price |
|---|---|---|---|---|---|---|---|
| RTX 5070 Ti | 16 Go | 81,5 tok/s | 3 817 | 9,6 Go | 179 W | 0,15 €/M tok | — |
| RTX 4070 Ti | 12 Go | 51,1 tok/s | 3 266 | 9,1 Go | 166 W | 0,23 €/M tok | — |
Not measured on these cards yet — click for an estimate:
Technical sheet
- Published
- 2025-01-20
- Architecture
- 48 layers · 8 KV heads · dimension 128
- Attention
- Standard: every layer keeps the whole context in memory.
- Context cache (f16)
- ≈ 188 MB per 1,000 tokens · 6,0 Go for 32,768 tokens
- GGUF file
- bartowski/DeepSeek-R1-Distill-Qwen-14B-GGUF · DeepSeek-R1-Distill-Qwen-14B-Q4_K_M.gguf Download (8,4 Go)
- Official repo
- deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.