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

gpt-oss-120b

by openai

120B params · text-generation · 4.6k likes · 4.4M downloads

gpt-oss-120b is a 120B parameter model. At Q4 quantization it requires 60GB of VRAM. It runs comfortably on H100 SXM5 80 GB (40 tok/s), H200 SXM 141 GB (58 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
Groq435 tok/s
Novita51 tok/s
Cerebras206 tok/s
SambaNova552 tok/s
Nscale95 tok/s
Hyperbolic90 tok/s
Together AI81 tok/s
Fireworks90 tok/s
Scaleway116 tok/s
OVHcloud131 tok/s

GPU compatibility

GPUVRAMQ4 DecodeVerdict
M4 Max 128GB128GB3 tok/stight
A100 PCIe 80 GB80GB19 tok/stight
H100 SXM5 80 GB80GB40 tok/scomfortable
H200 SXM 141 GB141GB58 tok/scomfortable
M2 Ultra 192GB192GB5 tok/stight
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0