Configs › GLM-5.3-Flash tok/s: Mac Ultra 55.2, DGX Spark 38
GLM-5.3-Flash tok/s: Mac Ultra 55.2, DGX Spark 38
Measured GLM-5.3-Flash speeds: 55.2 tok/s on Mac Ultra (mlx (speculative decode, mtp)). 5 configs, each linked to its source, snapshot 2026-10-16.
Snapshot 2026-10-16 · 394 configs on AI boxes and 423 on GPU cards · 123 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
Quick answers
- How fast is GLM-5.3-Flash on Mac Ultra?
- 55.2 tok/s single-stream decode (mlx (speculative decode, mtp)), the fastest of 2 measured configs on Apple Mac Studio (M-series Ultra), measured by EugeneClaw, source last checked 2026-09-29. source
- How fast is GLM-5.3-Flash on DGX Spark?
- 38 tok/s single-stream decode (vllm), the fastest of 2 measured configs on NVIDIA DGX Spark (GB10), measured by gitcommit90, source last checked 2026-09-16. source
- How fast is GLM-5.3-Flash on Strix Halo?
- 30 tok/s single-stream decode (exllamav3 (speculative decode, spec)), the only measured config on AMD Strix Halo (Ryzen AI Max+ 395), measured by Yamz-Labs, source last checked 2026-10-05. source
Measured configs for GLM-5.3-Flash
| Model | Hardware | Quant | Backend | Decode | Context | Trust | Source |
|---|---|---|---|---|---|---|---|
| GLM-5.3-Flash | Mac Ultra | mlx (speculative decode, mtp) | 55.2 tok/s | Extracted, source-linked | source | ||
| GLM-5.3-Flash | DGX Spark | vllm | 38 tok/s | 262144 | Owner-submitted | source | |
| GLM-5.3-Flash | Mac Ultra | MLX-4BIT | mlx | 34.2 tok/s | Extracted, source-linked | source | |
| GLM-5.3-Flash | Strix Halo | exllamav3 (speculative decode, spec) | 30 tok/s (prefill 613) | 65536 | Extracted, source-linked | source | |
| GLM-5.3-Flash | DGX Spark | vllm | 25.1 tok/s | Structured table | source |
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.