Zhipu · 30B · MoE
Can you run GLM-4.7-Flash 30B-A3B locally?
GLM-4.7-Flash 30B-A3B is a mixture-of-experts model with 30B parameters, 3B active per token. At 4-bit or better it runs on 73 of the 120 hardware configurations listed here; the smallest is the RTX 4090 (24 GB). On a 128 GB DGX Spark it runs at Q4_K_M, at roughly 79 tokens per second (our est.).
Data as of 2 November 2026
Fastest machines for GLM-4.7-Flash 30B-A3B
Every machine that runs it, fastest first. Filter by type; your machine is marked. Compare machines running it side by side
- 1RTX 5090Graphics card · 32 GB~273 tok/s ModelFit est.Q4_K_M · 20 of 31 GB
- 2RTX PRO 6000 BlackwellGraphics card · 96 GB~273 tok/s ModelFit est.Q4_K_M · 20 of 95 GB
- 32x RTX 5090Graphics card · 64 GB~252 tok/s ModelFit est.Q4_K_M · 20 of 62 GB
- 42x RTX PRO 6000 BlackwellGraphics card · 192 GB~252 tok/s ModelFit est.Q4_K_M · 20 of 190 GB
- 5RTX 4090Graphics card · 24 GB~178 tok/s ModelFit est.Q4_K_M · 20 of 23 GB
- 6M5 UltraMac · 96–512 GB · 5 variants~175 tok/s ModelFit est.Q4_K_M · 20 GB
- 72x RTX 4090Graphics card · 48 GB~169 tok/s ModelFit est.Q4_K_M · 20 of 46 GB
- 8RTX 3090 TiGraphics card · 24 GB~165 tok/s ModelFit est.Q4_K_M · 20 of 23 GB
- 92x RTX 3090 TiGraphics card · 48 GB~157 tok/s ModelFit est.Q4_K_M · 20 of 46 GB
- 10RTX 3090Graphics card · 24 GB~156 tok/s ModelFit est.Q4_K_M · 20 of 23 GB
- 112x RTX 3090Graphics card · 48 GB~149 tok/s ModelFit est.Q4_K_M · 20 of 46 GB
- 12RX 7900 XTXGraphics card · 24 GB~132 tok/s ModelFit est.Q4_K_M · 20 of 23 GB
- 132x RX 7900 XTXGraphics card · 48 GB~127 tok/s ModelFit est.Q4_K_M · 20 of 46 GB
- 14M3 UltraMac · 96–512 GB · 5 variants~123 tok/s ModelFit est.Q4_K_M · 20 GB
- 15M5 MaxMac · 48–128 GB · 3 variants~116 tok/s ModelFit est.Q4_K_M · 20 GB
- 16M2 UltraMac · 64–192 GB · 7 variants~108 tok/s ModelFit est.Q4_K_M · 20 GB
- 17M4 MaxMac · 48–128 GB · 3 variants~97 tok/s ModelFit est.Q4_K_M · 20 GB
- 18M1 UltraMac · 64–128 GB · 4 variants~97 tok/s ModelFit est.Q4_K_M · 20 GB
- 19M5 Max · 36 GB · 32cMac · 36 GB~90 tok/s ModelFit est.Q4_K_M · 20 of 24 GB
- 20DGX Spark (GB10)AI mini PC · 128 GB~79 tok/s our est.Q4_K_M · 20 of 122 GB
- 21M3 MaxMac · 48–128 GB · 3 variants~76 tok/s ModelFit est.Q4_K_M · 20 GB
- 22M4 Max · 36 GB · 32cMac · 36 GB~76 tok/s ModelFit est.Q4_K_M · 20 of 24 GB
- 23Ryzen AI Max+ 395 · 64 GBAI mini PC · 64 GB~74 tok/s our est.Q4_K_M · 20 of 48 GB
- 24M2 MaxMac · 32–96 GB · 5 variants~71 tok/s ModelFit est.Q4_K_M · 20 GB
- 25M5 ProMac · 48–64 GB · 4 variants~68 tok/s ModelFit est.Q4_K_M · 20 GB
- 26M1 MaxMac · 32–64 GB · 4 variants~64 tok/s ModelFit est.Q4_K_M · 20 GB
- 27M3 MaxMac · 36–96 GB · 2 variants~59 tok/s ModelFit est.Q4_K_M · 20 GB
- 28M4 ProMac · 48–64 GB · 4 variants~57 tok/s ModelFit est.Q4_K_M · 20 GB
- 29Ryzen AI Max+ 395 · 128 GBAI mini PC · 128 GB~45 tok/s ModelFit est.Q4_K_M · 20 of 96 GB
- 30M2 Pro · 32 GBMac · 32 GB~42 tok/s ModelFit est.Q4_K_M · 20 of 21 GB
- 31M6 · 32 GB · 12cMac · 32 GB~40 tok/s ModelFit est.Q4_K_M · 20 of 21 GB
- 32M1 Pro · 32 GBMac · 32 GB~38 tok/s ModelFit est.Q4_K_M · 20 of 21 GB
- 33M5Mac · 32 GB · 2 variants~35 tok/s ModelFit est.Q4_K_M · 20 GB
- 34M3 Pro · 36 GBMac · 36 GB~34 tok/s ModelFit est.Q4_K_M · 20 of 24 GB
- 35M4 · 32 GBMac · 32 GB~26 tok/s ModelFit est.Q4_K_M · 20 of 21 GB
Too big for 47 machines
RX 7900 XT, M5 Pro · 24 GB · 20c, M5 Pro · 24 GB · 16c, M4 Pro · 24 GB · 20c, M4 Pro · 24 GB · 16c, M6 · 24 GB · 12c, M6 · 24 GB · 10c, M5 · 24 GB · 10c, M5 · 24 GB · 8c, M4 · 24 GB, M3 · 24 GB, M2 · 24 GB, RTX 5080, RTX 5070 Ti, RTX 5060 Ti, RTX 4080 SUPER, RTX 4080, RTX 4070 Ti SUPER, RTX 4060 Ti, Tesla T4, Tesla P100, RX 9070 XT, RX 9060 XT, M3 Pro · 18 GB, RTX 5070, RTX 4070 Ti, RTX 4070 SUPER, RTX 4070, RTX 3060, M2 Pro · 16 GB, M1 Pro · 16 GB, M6 · 16 GB · 12c, M6 · 16 GB · 10c, M5 · 16 GB · 10c, M5 · 16 GB · 8c, M4 · 16 GB, M3 · 16 GB, M2 · 16 GB, M1 · 16 GB, RTX 5060, RTX 4060, RTX 3070, RTX 3060 Ti, RTX 2070, M3 · 8 GB, M2 · 8 GB, M1 · 8 GB. GLM-4.7-Flash 30B-A3B needs about 20 GB at its smallest build.
Speeds are single-user decode estimates, not measurements. "ModelFit est." comes from ModelFit's model fitted on published runs (CC BY 4.0); "our est." is memory bandwidth divided by bytes read per token, times 0.55. Memory is what the best build needs, against what the machine can give a model. Method
About the model
- Parameters
- 30B
- Active per token
- 3B
- Type
- Mixture of experts
- Good for
- Coding, Agentic tasks, Reasoning
Builds and how to run them
| Quantization | Load | Min. RAM | Command |
|---|---|---|---|
| Q4_K_M | 19.0 GB | 32 GB | ollama run glm-4.7-flash |
Load, minimum RAM and Ollama commands: ModelFit, CC BY 4.0. Load is memory with the model loaded, before long contexts.
Questions
Can I run GLM-4.7-Flash 30B-A3B on a Mac?
Yes on a Mac with enough memory: the smallest Apple configuration listed here that runs it at 4-bit or better is the M2 Max · 32 GB · 38c.
How much memory does GLM-4.7-Flash 30B-A3B need?
About 20 GB at the smallest build in our data, including a margin for the runtime and an 8K-token context. Longer contexts need more.
Does GLM-4.7-Flash 30B-A3B run on a DGX Spark?
Yes, at Q4_K_M. The Spark has 128 GB of unified memory, about 121.6 GB usable.