RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run GLM 4.1V 9B Thinking locally?

GLM 4.1V 9B Thinking by zai-org has 10B parameters. At 4-bit with an 8K context it needs about 6.7 GB of GPU memory. It fits on 65 of 65 GPUs and Macs we track. See all models that fit in 8 GB of VRAM.

6.7 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest Mac that fits
64Kmax context
GQAattention type

Memory needed by quant and context length

GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from unsloth/GLM-4.1V-9B-Thinking-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx
2-bit (Q2_K) •rough4.6 GB4.7 GB5.7 GB
3-bit (Q3_K_M) •usable5.5 GB5.6 GB6.6 GB
4-bit (Q4_K_M) •good, most popular6.6 GB6.7 GB7.7 GB
5-bit (Q5_K_M) •very good7.4 GB7.6 GB8.5 GB
6-bit (Q6_K) •near lossless8.5 GB8.7 GB9.6 GB
8-bit (Q8_0) •effectively lossless10.2 GB10.3 GB11.3 GB
16-bit (F16/BF16) •original weights18.4 GB18.5 GB19.5 GB

Which GPUs and Macs run GLM 4.1V 9B Thinking?

Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).

HardwareMemoryFits?Best quantNeedsMax tok/s (est.)
RTX 3060 12GB12 GBFitsQ6_K8.7 GB44details
RTX 4060 Ti 16GB16 GBFitsQ8_010.3 GB29details
RTX 309024 GBFitsF1618.5 GB50details
RTX 409024 GBFitsF1618.5 GB54details
RTX 509032 GBFitsF1618.5 GB95details
RTX 5070 Ti16 GBFitsQ8_010.3 GB90details
RX 7900 XTX24 GBFitsF1618.5 GB51details
Mac M4 Pro 48GB48 GBFitsF1618.5 GB15details
Mac M4 Max 128GB128 GBFitsF1618.5 GB29details
Mac M3 Ultra 512GB512 GBFitsF1618.5 GB44details
DGX Spark128 GBFitsF1618.5 GB15details
RTX 407012 GBFitsQ6_K8.7 GB61details
RTX 5060 Ti 16GB16 GBFitsQ8_010.3 GB45details
RTX 40608 GBFitsQ4_K_M6.7 GB44details
Arc B58012 GBFitsQ6_K8.7 GB55details
RTX 3060 Ti8 GBFitsQ4_K_M6.7 GB73details
RTX 30708 GBFitsQ4_K_M6.7 GB73details
RTX 3080 10GB10 GBFitsQ5_K_M7.6 GB108details
RTX 3080 12GB12 GBFitsQ6_K8.7 GB110details
RTX 3090 Ti24 GBFitsF1618.5 GB54details
RTX 4060 Ti 8GB8 GBFitsQ4_K_M6.7 GB47details
RTX 4070 Super12 GBFitsQ6_K8.7 GB61details
RTX 4070 Ti12 GBFitsQ6_K8.7 GB61details
RTX 4070 Ti Super16 GBFitsQ8_010.3 GB67details
RTX 408016 GBFitsQ8_010.3 GB72details
RTX 4080 Super16 GBFitsQ8_010.3 GB74details
RTX 50608 GBFitsQ4_K_M6.7 GB73details
RTX 507012 GBFitsQ6_K8.7 GB81details
RTX 508016 GBFitsQ8_010.3 GB96details
2x RTX 30902×24 GBFitsF1618.5 GB50details
2x RTX 40902×24 GBFitsF1618.5 GB54details
2x RTX 50902×32 GBFitsF1618.5 GB95details
RTX A600048 GBFitsF1618.5 GB41details
RTX 6000 Ada48 GBFitsF1618.5 GB51details
RTX PRO 6000 Blackwell96 GBFitsF1618.5 GB95details
L40S48 GBFitsF1618.5 GB46details
A100 40GB40 GBFitsF1618.5 GB83details
A100 80GB80 GBFitsF1618.5 GB108details
H100 PCIe80 GBFitsF1618.5 GB106details
H100 SXM80 GBFitsF1618.5 GB178details
H200141 GBFitsF1618.5 GB255details
RX 76008 GBFitsQ4_K_M6.7 GB47details
RX 7700 XT12 GBFitsQ6_K8.7 GB52details
RX 7800 XT16 GBFitsQ8_010.3 GB62details
RX 7900 GRE16 GBFitsQ8_010.3 GB58details
RX 7900 XT20 GBFitsQ8_010.3 GB80details
RX 907016 GBFitsQ8_010.3 GB64details
RX 9070 XT16 GBFitsQ8_010.3 GB64details
Ryzen AI Max+ 395 128GB128 GBFitsF1618.5 GB14details
Arc A770 16GB16 GBFitsQ8_010.3 GB56details
Mac M1 16GB16 GBFitsQ6_K8.7 GB8.2details
Mac M2 24GB24 GBFitsQ8_010.3 GB10details
Mac M3 24GB24 GBFitsQ8_010.3 GB10details
Mac M4 16GB16 GBFitsQ6_K8.7 GB15details
Mac M4 32GB32 GBFitsF1618.5 GB6.4details
Mac M1 Pro 32GB32 GBFitsF1618.5 GB11details
Mac M3 Pro 36GB36 GBFitsF1618.5 GB8.0details
Mac M4 Pro 24GB24 GBFitsQ8_010.3 GB27details
Mac M4 Pro 64GB64 GBFitsF1618.5 GB15details
Mac M1 Max 64GB64 GBFitsF1618.5 GB21details
Mac M2 Max 96GB96 GBFitsF1618.5 GB21details
Mac M3 Max 128GB128 GBFitsF1618.5 GB21details
Mac M4 Max 64GB64 GBFitsF1618.5 GB29details
Mac M2 Ultra 192GB192 GBFitsF1618.5 GB43details
Mac M3 Ultra 256GB256 GBFitsF1618.5 GB44details

Questions

How much VRAM does GLM 4.1V 9B Thinking need?

About 6.7 GB at 4-bit (Q4_K_M) with an 8K context: 5.7 GB of weights (real GGUF file size), 0.3 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 10.3 GB.

What is the smallest GPU that runs GLM 4.1V 9B Thinking?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 4060 (8 GB).

Can I run GLM 4.1V 9B Thinking on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M1 16GB or larger, using the default macOS GPU memory limit.

Similar models

Same family or a similar parameter count, with the VRAM each needs at Q4_K_M and 8K context.

GLM 4.6V Flash

10B · 6.7 GB at 4-bit

glm 4 9b

9.4B · 11.5 GB at 4-bit

GLM 4.7 Flash

31B · 18.2 GB at 4-bit

Step3 VL 10B

10B · 6.5 GB at 4-bit

Llama 3.2 11B Vision

11B · 8.0 GB at 4-bit

Llama 3.2 11B Vision Instruct

11B · 7.4 GB at 4-bit

Model facts

Hugging Face repozai-org/GLM-4.1V-9B-Thinking
Parameters10,292,777,472 (10B)
Layers / hidden size40 / 4096
AttentionGQA; 32 heads, 2 KV heads, head dim 128
KV cache per token (FP16)40.0 KiB
Max context65,536 tokens
Licensemit
Released on HFJun 28, 2025
Downloads (30d) / likes40,523 / 792

Sources: Hugging Face model API · config.json (zai-org/GLM-4.1V-9B-Thinking) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.