←back to thread

Gemma 3 QAT Models: Bringing AI to Consumer GPUs

(developers.googleblog.com)
602 points emrah | 1 comments | 20 Apr 25 12:22 UTC | HN request time: 0.204s | source
Show context
holografix ◴[20 Apr 25 13:27 UTC] No.43743631[source]▶
Could 16gb vram be enough for the 27b QAT version?
replies(5): >>43743634 #>>43743704 #>>43743825 #>>43744249 #>>43756253 #
jffry ◴[20 Apr 25 13:39 UTC] No.43743704[source]▶
With `ollama run gemma3:27b-it-qat "What is blue"`, GPU memory usage is just a hair over 20GB, so no, probably not without a nerfed context window
replies(1): >>43743804 #
1. woadwarrior01 ◴[20 Apr 25 13:58 UTC] No.43743804[source]▶
Indeed, the default context length in ollama is a mere 2048 tokens.