RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run granite 4.2 30b locally?

granite 4.2 30b by ibm-granite has 29B parameters. At 4-bit with an 8K context it needs about 19.1 GB of GPU memory. It fits on 48 of 65 GPUs and Macs we track. See all models that fit in 24 GB of VRAM.

19.1 GBQ4_K_M · 8K context
RTX 3090smallest consumer GPU that fits
M4 32GBsmallest Mac that fits
128Kmax 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 ibm-granite/granite-4.2-30b-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K) •rough11.7 GB12.7 GB18.7 GB42.7 GB
3-bit (Q3_K_M) •usable14.8 GB15.8 GB21.8 GB45.8 GB
4-bit (Q4_K_M) •good, most popular18.1 GB19.1 GB25.1 GB49.1 GB
5-bit (Q5_K_M) •very good20.9 GB21.9 GB27.9 GB51.9 GB
6-bit (Q6_K) •near lossless24.0 GB25.0 GB31.0 GB55.0 GB
8-bit (Q8_0) •effectively lossless30.6 GB31.6 GB37.6 GB61.6 GB
16-bit (F16/BF16) •original weights56.1 GB57.1 GB63.1 GB87.1 GB

Which GPUs and Macs run granite 4.2 30b?

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 GBNo———details
RTX 4060 Ti 16GB16 GBFitsQ2_K12.7 GB27details
RTX 309024 GBFitsQ4_K_M19.1 GB53details
RTX 409024 GBFitsQ4_K_M19.1 GB57details
RTX 509032 GBFitsQ6_K25.0 GB75details
RTX 5070 Ti16 GBFitsQ2_K12.7 GB83details
RX 7900 XTX24 GBFitsQ4_K_M19.1 GB54details
Mac M4 Pro 48GB48 GBFitsQ6_K25.0 GB11details
Mac M4 Max 128GB128 GBFitsF1657.1 GB9.3details
Mac M3 Ultra 512GB512 GBFitsF1657.1 GB14details
DGX Spark128 GBFitsF1657.1 GB4.7details
RTX 407012 GBNo———details
RTX 5060 Ti 16GB16 GBFitsQ2_K12.7 GB41details
RTX 40608 GBNo———details
Arc B58012 GBNo———details
RTX 3060 Ti8 GBNo———details
RTX 30708 GBNo———details
RTX 3080 10GB10 GBNo———details
RTX 3080 12GB12 GBNo———details
RTX 3090 Ti24 GBFitsQ4_K_M19.1 GB57details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBNo———details
RTX 4070 Ti12 GBNo———details
RTX 4070 Ti Super16 GBFitsQ2_K12.7 GB62details
RTX 408016 GBFitsQ2_K12.7 GB66details
RTX 4080 Super16 GBFitsQ2_K12.7 GB68details
RTX 50608 GBNo———details
RTX 507012 GBNo———details
RTX 508016 GBFitsQ2_K12.7 GB88details
2x RTX 30902×24 GBFitsQ8_031.6 GB30details
2x RTX 40902×24 GBFitsQ8_031.6 GB32details
2x RTX 50902×32 GBFitsQ8_031.6 GB58details
RTX A600048 GBFitsQ8_031.6 GB25details
RTX 6000 Ada48 GBFitsQ8_031.6 GB31details
RTX PRO 6000 Blackwell96 GBFitsF1657.1 GB31details
L40S48 GBFitsQ8_031.6 GB28details
A100 40GB40 GBFitsQ8_031.6 GB50details
A100 80GB80 GBFitsF1657.1 GB35details
H100 PCIe80 GBFitsF1657.1 GB34details
H100 SXM80 GBFitsF1657.1 GB57details
H200141 GBFitsF1657.1 GB82details
RX 76008 GBNo———details
RX 7700 XT12 GBNo———details
RX 7800 XT16 GBFitsQ2_K12.7 GB57details
RX 7900 GRE16 GBFitsQ2_K12.7 GB53details
RX 7900 XT20 GBFitsQ3_K_M15.8 GB57details
RX 907016 GBFitsQ2_K12.7 GB59details
RX 9070 XT16 GBFitsQ2_K12.7 GB59details
Ryzen AI Max+ 395 128GB128 GBFitsF1657.1 GB4.4details
Arc A770 16GB16 GBFitsQ2_K12.7 GB52details
Mac M1 16GB16 GBNo———details
Mac M2 24GB24 GBFitsQ2_K12.7 GB9.2details
Mac M3 24GB24 GBFitsQ2_K12.7 GB9.2details
Mac M4 16GB16 GBNo———details
Mac M4 32GB32 GBFitsQ4_K_M19.1 GB6.8details
Mac M1 Pro 32GB32 GBFitsQ4_K_M19.1 GB11details
Mac M3 Pro 36GB36 GBFitsQ4_K_M19.1 GB8.5details
Mac M4 Pro 24GB24 GBFitsQ2_K12.7 GB25details
Mac M4 Pro 64GB64 GBFitsQ8_031.6 GB8.8details
Mac M1 Max 64GB64 GBFitsQ8_031.6 GB13details
Mac M2 Max 96GB96 GBFitsF1657.1 GB6.8details
Mac M3 Max 128GB128 GBFitsF1657.1 GB6.8details
Mac M4 Max 64GB64 GBFitsQ8_031.6 GB18details
Mac M2 Ultra 192GB192 GBFitsF1657.1 GB14details
Mac M3 Ultra 256GB256 GBFitsF1657.1 GB14details

Questions

How much VRAM does granite 4.2 30b need?

About 19.1 GB at 4-bit (Q4_K_M) with an 8K context: 16.5 GB of weights (real GGUF file size), 2.0 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 31.6 GB.

What is the smallest GPU that runs granite 4.2 30b?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 3090 (24 GB).

Can I run granite 4.2 30b on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M4 32GB 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.

granite 4.0 h small

32B · 18.9 GB at 4-bit

granite 4.2 8b

8.8B · 6.8 GB at 4-bit

granite 4.1 8b

8.8B · 6.8 GB at 4-bit

Muse Glimmer 30B

30B · 17.9 GB at 4-bit

Hy Memory 30B A3B

30B · 18.5 GB at 4-bit

North Mini Code 1.0

30B · 19.3 GB at 4-bit

Model facts

Hugging Face repoibm-granite/granite-4.2-30b
Parameters29,276,770,304 (29B)
Layers / hidden size64 / 4096
AttentionGQA; 32 heads, 8 KV heads, head dim 128
KV cache per token (FP16)256.0 KiB
Max context131,072 tokens
Licenseapache-2.0
Released on HFAug 7, 2026
Downloads (30d) / likes37,188 / 143

Sources: Hugging Face model API · config.json (ibm-granite/granite-4.2-30b) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.