Model benchmarks

zai-org/glm-4.7-flash local LLM performance

As of March 2026, zai-org/glm-4.7-flash runs at up to 46.1 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).

LM Studio

Model size

Unknown

Peak speed

46.1 tok/s

Average speed

45.6 tok/s

Min memory

n/a

Max context

32,768 tokens

Best quality

78.0

Benchmark runs

3

GPUs tested

1

Performance by hardware and tool

Every hardware/tool/quantization combination zai-org/glm-4.7-flash has been benchmarked on, ranked by peak token generation speed. Last updated March 2026.

HardwareToolQuantBest tok/sAvg tok/sMemoryContextQualityRuns
AMD Radeon RX 7900 XTXLM Studio46.1 tok/s45.6 tok/sn/a32,768 tokens47.43

Frequently asked questions

How fast is zai-org/glm-4.7-flash for local inference?
Across 3 community benchmark runs, zai-org/glm-4.7-flash reaches up to 46.1 tok/s and averages 45.6 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
How much memory does zai-org/glm-4.7-flash need?
The leanest observed configuration used about n/a of memory.
Which tools have been used to run zai-org/glm-4.7-flash?
Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.