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

command-a-reasoning-08-2025

by CohereLabs

111B params · text-generation · 134 likes · 1.0k downloads

command-a-reasoning-08-2025 is a 111B parameter model. At Q4 quantization it requires 56GB of VRAM. It runs comfortably on H100 SXM5 80 GB (43 tok/s), H200 SXM 141 GB (63 tok/s).

Inference providers

Provider$/1M in$/1M outThroughput
Cohere88 tok/s

GPU compatibility

GPUVRAMQ4 DecodeVerdict
M4 Max 128GB128GB4 tok/stight
A100 PCIe 80 GB80GB21 tok/stight
H100 SXM5 80 GB80GB43 tok/scomfortable
H200 SXM 141 GB141GB63 tok/scomfortable
M2 Ultra 192GB192GB6 tok/stight
Install CLI [email protected] Raw data · MIT · API data: live · HW/Cloud data: curated 2026-02-23 · v0.6.0