今天想和大家分享一个8G显存跑视频工作流的最后一块版图，完全基于 Qwen3.8-27B 模型架构的开源权重 Ternary Bonsai 2 27B
This is a AI post classified by Jev as Open & local models (a model release), kept by the AI Radar because it carries real work, not commentary.
今天想和大家分享一个8G显存跑视频工作流的最后一块版图，完全基于 Qwen3.8-27B 模型架构的开源权重 Ternary Bonsai 2 27B Bonsai 2 27B 最大的亮点，是把约 54GB 的 Qwen3.8-27B 压到 5.95GB（缩小了9倍） 同时官方测试仍保留 98.2% 的综合能力，尤其数学、编程和工具调用损失很小 在英伟达RTX 5090 可以达到 143 tok/s，在Mac M5 Max上可以达到 46.8 tok/s 这有可能是最适合16G显存 甚至8G显存的本地模型 Bonsai 2 27B 写剧本拆分镜 -> Qwen-Image-2.1 生成关键帧 -> MiniMax H3 生成视频
Posted by 雪瑜 (5.9k followers) 1 h ago · 5 likes · 465 views · view the original post on X. Kept by the AI Radar as Open & local models.
More AI work like this
- NEW UNCENSORED Xiaomi MiMo-V2.6-Flash model you can run locally with Zero refusals. — @0x0SojalSec
- these are the best cheap models right now — @nicdunz
- Qwen-Image-2.1 开放权重后，许可证成了最大的争议点 — @xueyu1125
- Locally Laya just beat cloud Jev at live Tetris. — @0x0SojalSec
- Meet Qwen-Image-2.1-Text-Encoder-Heretic-GGUF: a quantized text encoder that powers Qwen… — @HuggingModels
- Jev-Omni the first open multimodal “system 1” model. — @0x0SojalSec
- MiMo-V2.6-Pro is live on Wally. Day 1. — @RunAnywhereAI
- Nano Banana 2.0 level Qwen-Image-2.1 now runs locally on 4.6 GB laptops no filter. — @0x0SojalSec
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 43.7k posts from 5k X accounts over the last 14 days, 1.8k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-23 02:25 UTC. Full method.