GLM 4.7 Flash
A lightweight programming-oriented model in Zhipu AI's GLM series, designed to support local deployment. The model file is approximately 19 GB in size.
Timeline (last 90 days)
64GB of memory can run two GLM 4.7 Flash instances in parallel, which the author's other PC, with higher specifications but less memory, cannot do.
In the 16-concurrency, 256-token-output test, Gemma 4 E2B achieves an output speed of 1887 tokens/second, and the models faster than it include Qwen2.5-0.5B (4176, 2.21x), Qwen3-0.6B (3210, 1.70x), Qwen2.5-1.5B (2780, 1.47x), and Qwen3-1.7B (2049, 1.09x).
GLM 4.7 Flash is licensed under MIT and scored 59.2 on SWE-Bench Verified.
All Facts (3)
64GB of memory can run two GLM 4.7 Flash instances in parallel, which the author's other PC, with higher specifications but less memory, cannot do.
50%UnverifiedIn the 16-concurrency, 256-token-output test, Gemma 4 E2B achieves an output speed of 1887 tokens/second, and the models faster than it include Qwen2.5-0.5B (4176, 2.21x), Qwen3-0.6B (3210, 1.70x), Qwen2.5-1.5B (2780, 1.47x), and Qwen3-1.7B (2049, 1.09x).
50%UnverifiedGLM 4.7 Flash is licensed under MIT and scored 59.2 on SWE-Bench Verified.
50%