vram.run Models Hardware Providers Cloud State of Inference
API provider data is live · Hardware & cloud pricing curated 2026-02-23

Groq vs Nscale

8 vs 24 models, 5 shared

Shared models

ModelGroq $/1M outGroq tok/sNscale $/1M outNscale tok/s
Qwen3-32B302 tok/s27 tok/s
Llama-3.3-70B-Instruct295 tok/s16 tok/s
Llama-4-Scout-17B-16E-Instruct338 tok/s32 tok/s
gpt-oss-120b435 tok/s95 tok/s
gpt-oss-20b558 tok/s121 tok/s
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0