Edge0 releases Edge0-145B and Edge0-120B, open-weight models it says run on a laptop

Edge0 released the weights of Edge0-145B and Edge0-120B on November 12, together with its inference framework and deployment tools. The company calls Edge0-145B the first open-source on-device model with 145 billion parameters (its claim).

• Edge0 says the model is a sparse mixture of experts that uses about 10B of its 145B parameters per step, with groups of 64 experts sharing two large matrices. It reports about 29GB of peak memory on a laptop, 80+ tokens/s decoding, 1,100+ tokens/s prompt processing, about 120W of power and a 128K–256K context.
• On Edge0's own evaluations, it scores 64.6 on DeepSWE 1.1, 86.52 on Terminal-Bench 2.1, 90.4 on GPQA-Diamond and 80.6 on IFBench. Edge0 puts these ahead of or close to Qwen3.8-Flash-Next, DeepSeek-V4-Flash-0731 and Claude Opus 4.6 (Max) on the benchmarks it names. The full evaluation setup is due with a technical report.
• Edge0 says day-one support covers macOS, Windows and Linux, including Apple M-series chips and NVIDIA RTX GPUs. It shows the model running local agent workflows in Blender, KiCad and drone control.