llamaperf
Calculator
GPUs
Engines
Guides
Blog
MCP
Advertise
Submit
Sign in
Submit
llamaperf
/
NVIDIA RTX Pro 6000 Blackwell
/
report
DeepSeek V4 Flash 284B (13B active)
on 2×
NVIDIA RTX Pro 6000 Blackwell
· vLLM
Tone: positive
NVIDIA hardware
Aug 28, 2026
VRAM reported
96 GB
Use cases
coding
Summary
User reports DeepSeek V4 Flash on 2x RTX PRO 6000 with vLLM reaches Sonnet quality and faster wall-clock time than Sonnet or Opus via API. User mentions Qwen 3.6 models as anchors.
View source ↗
Feedback