The letter

One letter a week, starting soon.

The week's signal, what shipped, what is worth running tonight, and the editor's note. No tracking, no ads, nothing else.

We keep your address, your language and the date you joined, nothing else. Every letter has a one-click unsubscribe link that deletes the record.

← Accept All   Archive
Qiita (LLM)Japan

24GB GPU(RTX 3090/4090/RX 7900 XTX)でローカルLLMは何が動く?数字で確認

11.04

4Kコンテキストで、129モデル中97が24GBに快適に収まり、102が一応ロード可能。 同じ24GBでも速度は帯域次第:Tesla P40の346 GB/sからRTX 4090の1,008 GB/sまで。生成速度はおおよそ「帯域 ÷ 重みファイルサイズ」。 Qwen3....

Chips & compute
Read at Qiita (LLM) ↗

Related

More from Qiita (LLM) on Accept All.