DeepSeek
R1 0528 (Qwen3) 8B
reasoning
8,2B
Parameters
4,7 Go
Q4_K_M file
≥ 6,2 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 4070 Ti | 12 Go | 90,1 tok/s | 5 994 | 5,2 Go | 154 W | 0,12 €/M tok | — |
Not measured on these cards yet — click for an estimate:
Technical sheet
- Published
- 2025-05-29
- Architecture
- 36 layers · 8 KV heads · dimension 128
- Attention
- Standard: every layer keeps the whole context in memory.
- Context cache (f16)
- ≈ 141 MB per 1,000 tokens · 4,5 Go for 32,768 tokens
- GGUF file
- bartowski/deepseek-ai_DeepSeek-R1-0528-Qwen3-8B-GGUF · deepseek-ai_DeepSeek-R1-0528-Qwen3-8B-Q4_K_M.gguf Download (4,7 Go)
- Official repo
- deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.