RTX PRO 6000 Blackwell Qwen3-8B Q4 token-gen @4K
174.0 tok/s TG batch-1 @4K ctx
Fields
id
benchmarks/comm-rtx-pro-6000-qwen3-8b-q4-tg-4kMetric
decode_tok_s
Value
174
Unit
tok/s TG batch-1 @4K ctx
Model
Qwen3-8B
Methodology
llama.cpp llama-bench, Q4 GGUF, CUDA, batch-1 decode (Hardware Corner measured table)
Measured by
third_party
Thermal state
unknown
notes
140 @16K, 111 @32K; 14B 96.9 @16K avg; Qwen3-32B 54.9 @4K / 23 @128K.
Notes
140 @16K, 111 @32K; 14B 96.9 @16K avg; Qwen3-32B 54.9 @4K / 23 @128K.