llama.cppで起動したMiMo-V2.6-Distill-Qwen-9Bにやらせた。ラーメン屋ベンチの結果です。動画をご査収ください。
This is a dev post classified by Jev as AI dev tools (a tool drop), kept by the Dev Radar because it carries real work, not commentary.
llama.cppで起動したMiMo-V2.6-Distill-Qwen-9Bにやらせた。ラーメン屋ベンチの結果です。動画をご査収ください。 あり得ない速度で、Q4_K_M GGUFに量子化されてるやつあったので、それでやりましたん。 リプライに貼っとくよぉ～。 Q4量子化後→5.63 GB<軽ｽｷﾞｨ MCode画面表示: 約 93.8 tok/s llama-server単独生成: 約 115〜119 tok/s 2ジョブ同時実行時: 約 94〜95 tok/s 画像はPixabay API渡して取ってこさせてます。 GPU: NVIDIA GeForce RTX 4090 24GB CPU: AMD Ryzen 9 7950X3D メモリ: 64GB モデル: MiMo-V2.6-Distill-Qwen-9B Q4_K_M GGUF ハーネス：Minimax Code
Posted by StudioYebisu (2.1k followers) 1 days ago · 87 likes · 9.5k views · view the original post on X. Kept by the Dev Radar as AI dev tools. Tools mentioned: mimo-v2.6-distill-qwen-9b.
More dev work like this
- Asked Claude Opus 5.5 to animate @PingRoomIO — @MahdiSPHP
- My ai-memory has more features than just sharing memory across agents. For example: it… — @AkitaOnRails
- Your coding agent can have a desktop sidekick — without a cloud account. — @DanKornas
- 🚨LATEST: OpenAI launches GPT-6 Sol and GPT-6 Luna with 50% lower API costs. — @coinbureau
- The news is out: DeepSeek-V4.1-Flash is free to use on WorkBuddy until October 9th! — @WorkBuddy_AI
- 斯坦福 2025 年秋季，开了一门 CS146S「现代软件开发者」，讲怎么跟编码 Agent 一起写软件，8 周作业全部公开在 GitHub 上。 — @GitHub_Daily
- @peterramsing @typesafeai Nice! I also built a Jev based ticket app. — @airesearch12
- in Executor v2, it's actually taking advantage of effect's error system now — @RhysSullivan
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. 22k posts from 5k X accounts over the last 21 days, 2.5k tools, 12 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-23 04:56 UTC. Full method.