APIUp to 28% cheaper than official pricesTry the API →
Hermes
APIModelsCases
Back to News
modelLlmsGpus

IsValorum Squeezes a 35B Open-Source Qwen3.8 Model Into 14.65 GB

A 35B Qwen mixture-of-experts release strips refusal vectors and squeezes the whole model plus 256K context into 24GB of VRAM.

IsValorum Squeezes a 35B Open-Source Qwen3.8 Model Into 14.65 GB
Source
AlphaSignal
Published
Author
AlphaSignal Newsroom
Read
1 min read

A 35B Qwen mixture-of-experts release strips refusal vectors and squeezes the whole model plus 256K context into 24GB of VRAM.

Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.

Read original report

Next reads

Prism ML Squeezes a 2B Vision Model Into 0.52 GB
PrismML · model

Prism ML Squeezes a 2B Vision Model Into 0.52 GB

Prism ML's Ternary-Bonsai-2 Shrinks a 9B AI Model to 2.42 GB
PrismML · model

Prism ML's Ternary-Bonsai-2 Shrinks a 9B AI Model to 2.42 GB

Unofficial Qwen3.8-27B Retrofit Runs 2.92x Faster at 261K Tokens
AlphaSignal · model

Unofficial Qwen3.8-27B Retrofit Runs 2.92x Faster at 261K Tokens