Skip to content
llm-speed

RTX 5090 vs Cerebras · Llama-3.3-70B

Side-by-side decode tok/s, prefill tok/s, and TTFT for RTX 5090 and Cerebras · Llama-3.3-70B, sourced from community-submitted runs of the llm-speed suite. Every number on this page links back to the run it came from.

hardware
RTX 5090
NVIDIA Ada/Blackwell · 32 GB VRAM
View RTX 5090 page →
hosted
Cerebras · Llama-3.3-70B
Cerebras · llama-3.3-70b

No overlapping RTX 5090 ↔ Cerebras · Llama-3.3-70B benchmarks yet.

We don't have a submitted run that covers both sides of this comparison yet. Run the suite on either side to populate this page:

$ pipx install llm-speed && llm-speed bench

See also: RTX 5090 benchmarks · Cerebras · Llama-3.3-70B · All hardware · All models · Methodology