RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can GLM 4.1V 9B Thinking run on the Mac M3 Ultra 512GB?

Yes. At 8K context the best fit is F16, using about 18.5 GB of the 384 GB usable on the Mac M3 Ultra 512GB. Estimated speed ceiling: ~44 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •4.6 GB Fits4.7 GB Fits5.7 GB Fits204
3-bit (Q3_K_M) •5.5 GB Fits5.6 GB Fits6.6 GB Fits165
4-bit (Q4_K_M) •6.6 GB Fits6.7 GB Fits7.7 GB Fits133
5-bit (Q5_K_M) •7.4 GB Fits7.6 GB Fits8.5 GB Fits116
6-bit (Q6_K) •8.5 GB Fits8.7 GB Fits9.6 GB Fits99
8-bit (Q8_0) •10.2 GB Fits10.3 GB Fits11.3 GB Fits82
16-bit (F16/BF16) •18.4 GB Fits18.5 GB Fits19.5 GB Fits44

• real GGUF file size from unsloth/GLM-4.1V-9B-Thinking-GGUF. Speed is a bandwidth ceiling (819 GB/s), not a benchmark. Methodology.

More

GLM 4.1V 9B Thinking on every GPU and Mac · Every model that runs on the Mac M3 Ultra 512GB · Apple Silicon guide · Try other settings in the calculator

Data: Hugging Face Hub (zai-org/GLM-4.1V-9B-Thinking), Apple specs. Data as of 2026-11-17.