Benchmark result
GLM-4.7-Flash — 90.5 tok/s
Measured with cpa manual on September 27, 2026.
What the model built
Open full screen →The coding scenario asks for a playable game in a single HTML file. This is exactly what the model returned, unedited; the frame only adds a line that tells this page how tall it is.
How this run compares
GLM-4.7-Flash on other hardware
- 90.5tok/s69.2Unknown hardwarethis run
- 50.7tok/s16.7Apple M4 Max
- 7.9tok/s69.2AMD Radeon RX 7900 XTX · Q4_K_M
- 6.1tok/s63.2AMD Radeon RX 7900 XTX · Q4_K_M
Reproduce this run
toolcpa manual
modelGLM-4.7-Flash
context8,192
clientv0.4.80+98
Set those in cpa, then:
llm-benchmark benchmark --model "GLM-4.7-Flash" --tool "cpa"The client prompts for context length and thinking mode, and for the KV cache dtype on Unsloth Studio. Sampling is left at the model default — the values above are what the tool reported using, not overrides the client sent.