DeepSeek Coder V2 Instruct (HF)

236B total / 21B active · load estimate uses total • 143.6GB VRAM (Q4 estimate) • 128,000 context

Estimate · Q4_K_M · 4k context · batch 1. For dense 7B–70B this sits ~5–8% above the Q4_K_M GGUF file. Not peak runtime VRAM and not a measured bench.