this is insane, my bonsai 2 27b build is trending on hugging face, first page of text generation, with 20,964 downloads in its first week. i…

this is insane, my bonsai 2 27b build is trending on hugging face, first page of text generation, with 20,964 downloads in its first week. it's prismml's ternary bonsai 2 with the qwen 3.8 mtp head grafted back on, packaged for the cards gamers actually own, prebuilt bundles for rtx 30, 40 and 50s and one serve script per card size, no compiling: > 8gb, a contributor's 3060 ti at 42.8 tok/s with 98k context > 12gb 3060, 50 tok/s with the mtp head > 16gb 5060 ti, 67 tok/s with the head, the full 262k window and vision loaded a 27b dense on an 8gb card, apache 2.0, running on your own gpu with nothing leaving it. drop your gpu and vram below and i'll point you to your four commands. my first model on huggingface has 11,792 downloads in 5 days, and it's a 27 billion parameter model doing 50 tok/s on a 5 year old rtx 3060. new here? it's bonsai 2 27b, qwen 3.8 27b compressed to ternary by prismml into one 7 gb file. it shipped without its mtp head, so i grafte…

Related

• Alert: Apple just dropped a new model on Hugging Face. It's a Qwen3.5-9B finetune that turns long documents into small page images to save t…
• 🤯QWEN 3.6 35B-A3B IS INSANE AT CODING @unslothai recently dropped Qwen3.6-35B-A3B-GGUF The first open Qwen3.6 model and it’s solid. This Mo…
• 180 tok/s generation on a 4090 with qwen 3.6. if you're on a 4090 and not running this model yet you're leaving performance on the table. 3B…

Source: Clem Delangue (X) | 2026-09-29