PrismML just put a 27B multimodal model on a phone. First time anyone's done it.
Bonsai 27B, built on Qwen3.6 27B with end-to-end ultra-low quantization.
Ternary: 5.9GB, 80.49 avg. Keeps 95% of FP16.
1-bit: 3.9GB, 76.11 avg. Math hits 93.40 — 2 points off FP16.
iPhone 17 Pro: 11 tok/s, 262K context.
A 27B model running locally on a phone. Wild.
PrismML just put a 27B multimodal model on a phone
More posts
kimi web ( kimi.com ) just added the k3
subscription-based skills become a real revenue model.
In my opinion, 2026 is year zero for general-purpose agents. once everyone's running one, subscription-based skills become a real revenue model. saas today is built for humans. the next wave is saas wrapped as an api skill that agents subscribe to on their own.
Meituan shipped LongCat-2.5
. 1.6T params, ~48B active, native multimodal, aimed at long-horizon agents, trained start to finish on Chinese hardware. Images get quantized into multi-level discrete tokens, and that's where Visual Coding and 3D pick up. On the RL side, failed trajectories get turned back into verifiable training tasks. W4A4 inference, 43% faster prefill. Meituan's own evals put it between Claude Opus 4.8 and Opus 5.
SenseTime open-sourced SenseNova Loop-9B-Preview.
It stacks layers instead of parameters. The decoder has 32 layers, but layers 4 through 27 run twice with shared weights, so a forward pass does 56 layers of work while the checkpoint stays at 9B. Their own eval numbers: 71.38 on BIRD-Dev and 13.72 on AutomationBench, both above Qwen3.5-27B. Terminal Bench is still behind.
