So you say you can't localmxaxx? Think again.
This is a dev post classified by AGK Intelligence as AI dev tools (a tool drop), kept by the Dev Radar because it carries real work, not commentary.
So you say you can't localmxaxx? Think again. Check this out bc this should not be running at ~22 tps on this PC. Someone got a 122B coding model running on 1x 24GB Radeon RX 7900 XTX. These are the spec of the rest of the machine 🎮 RX 7900 XTX → 24GB VRAM 🧠 RAM → 32GB DDR4-2133 🖥️ CPU → i7-7700K, 4 cores 🔌 PCIe → 3.0 💾 NVMe SSD The model is Ornith-1.5-122B. The UD-IQ2_XXS quant alone is 42GB. So obviously it doesn't fit in the GPU. Instead, a patched Strata setup profiles which MoE experts the model actually uses and tiers them: 🔥 hottest experts → VRAM 🧠 next tier → system RAM 💾 rare
Posted by David Hendrickson (12.6k followers) 2 days ago · 18 likes · 1.5k views · view the original post on X. Kept by the Dev Radar as AI dev tools.
More dev work like this
- 长PDF要是拆成单页再往一块拼，跨页表格和阅读顺序十有八九会乱套，这事你八成也烦过。 — @bkdgiffug
- Local models on Apple Silicon finally feel like a product, not a weekend project. — @PixelRainbowNFT
- GPT 6.1 Terra max vs GPT 6.1 Sol max — @HarshithLucky3
- Asked at the Meetup: "Does sbx run mistral work?" — @ajeetsraina
- Most agents freeze once deployed. — @GoogleCloudTech
- Your $250 GPU can now run a 35B coding model at 50 tokens/sec. — @0x0SojalSec
- Next week we're showing how to build self improving products. — @spenserskates
- Seedance 2.5 US Draft mode is now available on fal — @fal
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.2k posts from 5k X accounts over the last 21 days, 8.8k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-10-30 22:12 UTC. Full method.