look anon! someone with an rtx 3060 ti 8gb just opened a pr on my repo running bonsai 2…
This is a AI post classified by Jev as Open & local models (a tool drop), kept by the AI Radar because it carries real work, not commentary.
look anon! someone with an rtx 3060 ti 8gb just opened a pr on my repo running bonsai 2 27b at 32 → 43 tok/s. he built the kernel branch himself on windows, measured both binaries on the same card in the same session. the ladder so far, every row a different owner on their own card, same 5.9 GB file: > rtx 3060 ti 8gb: 32.44 → 43.40 tok/s, and the full 262k context fits in 7.7 GB > rtx 3060 12gb: 26 → 40 tok/s, 50 tok/s with the mtp head grafted back, greedy output bit identical > rtx 3090: 59.7 → 75.8 tok/s, measured by the prismml maintainers on their own box > rtx 4070 12gb: 51 → 103 tok
Posted by Sudo su (36.3k followers) 1 h ago · 31 likes · 1.9k views · view the original post on X. Kept by the AI Radar as Open & local models. Tools mentioned: bonsai2-small-gpu.
More AI work like this
- No Guardrails Qwen 2.1 Image — @Tech2Wild
- Qwen Image 2.1 is now open weight at just 7B parameters!! — @plotarmordev
- Local ultra-dense 27B quant showdown - The results are in: — @PaulGugAI
- きしださんのjevを採用させていただいています — @Tono_Ken3
- Hermes local models nailed this habit right for me. Every catalog row shows green /… — @MichaelGannotti
- なぜこんなマイナーなのにいい感じのModelをHFからサーベイすることが出来るんだい、ブラザー。 — @studio_yebisu
- They told us the models were going away. So Pirate Face built the continuation plan. — @thepirateface
- Bonsai 2 27B PTQ1_0 is working on A20 Pro (18PM) — @imnotchalk
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. 34.7k posts from 4.9k X accounts over the last 14 days, 1.5k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-20 17:01 UTC. Full method.