Sunk Cost sunkcost.ai Data checked 2026-09-03

Can a Strix Halo Corsair AI Workstation 300, 64GB run local LLMs?

Yes — 27 of the 39 open models on this site fit in its 48 GB of usable memory, the strongest being Qwen3.8 27B. Whether that saves you money is a different question, and the answer is usually no.

Price$1,700
Memory64 GB, about 48 GB of it addressable by the GPU at 256 GB/s
Best model it runsQwen3.8 27B — Sonnet-class, 10 tok/s
Pay-back against the APIPays back in 22 years at 500k tokens a day

Run the numbers on this machine

What it runs

ModelSpeedClassGood atMemory
Qwen3.8 27BQ4_K_M 10 tok/s Sonnet-class 19 GB
Qwen3.6 27BQ4_K_M 10 tok/s Haiku-class 19 GB
Qwen3.6 35B-A3BQ4_K_M 56 tok/s Haiku-class 23 GB
Muse Glimmer 30BQ4_K_M 11 tok/s Haiku-class 18 GB
Gemma 4 26B-A4BQ4_K_M 41 tok/s Haiku-class 18 GB
Gemma 4 31B itQ4_K_M 7.4 tok/s Haiku-class 26 GB
Granite 4.2 30BQ4_K_M 7.3 tok/s Haiku-class 26 GB
GLM-4.7-FlashQ4_K_M 40 tok/s Haiku-class 20 GB
Gemma 4 12BQ4_K_M 24 tok/s Below every hosted tier 8.0 GB
Qwen3.5 9BQ4_K_M 28 tok/s Below every hosted tier 6.8 GB
Nemotron 3.5 Lightning 30B-A3BQ4_K_M 54 tok/s Below every hosted tier 26 GB
MiniCPM5 2BQ4_K_M 65 tok/s Below every hosted tier 3.0 GB

15 more fit; the calculator lists them all.

The specifics

Chip
Corsair AI Workstation 300 — Ryzen AI Max 385 · Radeon 8050S, 32 CU
Memory bandwidth
256 GB/s
Usable by the GPU
48 GB — Windows lets you dedicate up to 75% of RAM to the GPU; Corsair lists up to 48 GB VRAM for this configuration. Measured GPU bandwidth on Strix Halo is ~212-215 GB/s of the 256 GB/s theoretical, so estimates here run a little high.
Power under load
133 W (stand in) — No measurement for the Max 385 system; stand-in is the Max+ 395 Framework board's measured 133 W, which likely overstates this smaller GPU.
Availability
Corsair's price on 2026-09-16, out of stock. This tier uses the smaller Ryzen AI Max 385.
Sources
source 1, source 2, source 3