RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run granite 4.1 3b llmon preview locally?

granite 4.1 3b llmon preview by ibm-granite has 3.7B parameters. At 4-bit with an 8K context it needs about 3.3 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.

3.3 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest 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 liodon-ai/granite-4.1-3b-llmon-preview-imatrix-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K)rough2.2 GB2.5 GB4.4 GB11.9 GB
3-bit (Q3_K_M)usable2.6 GB2.9 GB4.8 GB12.3 GB
4-bit (Q4_K_M) •good, most popular3.0 GB3.3 GB5.2 GB12.7 GB
5-bit (Q5_K_M) •very good3.3 GB3.7 GB5.5 GB13.0 GB
6-bit (Q6_K) •near lossless3.7 GB4.0 GB5.9 GB13.4 GB
8-bit (Q8_0) •effectively lossless4.5 GB4.8 GB6.7 GB14.2 GB
16-bit (F16/BF16)original weights7.7 GB8.0 GB9.9 GB17.4 GB

Which GPUs and Macs run granite 4.1 3b llmon preview?

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 GBFitsF168.0 GB49details
RTX 4060 Ti 16GB16 GBFitsF168.0 GB39details
RTX 309024 GBFitsF168.0 GB128details
RTX 409024 GBFitsF168.0 GB138details
RTX 509032 GBFitsF168.0 GB245details
RTX 5070 Ti16 GBFitsF168.0 GB122details
RX 7900 XTX24 GBFitsF168.0 GB131details
Mac M4 Pro 48GB48 GBFitsF168.0 GB37details
Mac M4 Max 128GB128 GBFitsF168.0 GB75details
Mac M3 Ultra 512GB512 GBFitsF168.0 GB112details
DGX Spark128 GBFitsF168.0 GB37details
RTX 407012 GBFitsF168.0 GB69details
RTX 5060 Ti 16GB16 GBFitsF168.0 GB61details
RTX 40608 GBFitsQ8_04.8 GB70details
Arc B58012 GBFitsF168.0 GB62details
RTX 3060 Ti8 GBFitsQ8_04.8 GB115details
RTX 30708 GBFitsQ8_04.8 GB115details
RTX 3080 10GB10 GBFitsF168.0 GB104details
RTX 3080 12GB12 GBFitsF168.0 GB125details
RTX 3090 Ti24 GBFitsF168.0 GB138details
RTX 4060 Ti 8GB8 GBFitsQ8_04.8 GB74details
RTX 4070 Super12 GBFitsF168.0 GB69details
RTX 4070 Ti12 GBFitsF168.0 GB69details
RTX 4070 Ti Super16 GBFitsF168.0 GB92details
RTX 408016 GBFitsF168.0 GB98details
RTX 4080 Super16 GBFitsF168.0 GB101details
RTX 50608 GBFitsQ8_04.8 GB115details
RTX 507012 GBFitsF168.0 GB92details
RTX 508016 GBFitsF168.0 GB131details
2x RTX 30902×24 GBFitsF168.0 GB128details
2x RTX 40902×24 GBFitsF168.0 GB138details
2x RTX 50902×32 GBFitsF168.0 GB245details
RTX A600048 GBFitsF168.0 GB105details
RTX 6000 Ada48 GBFitsF168.0 GB131details
RTX PRO 6000 Blackwell96 GBFitsF168.0 GB245details
L40S48 GBFitsF168.0 GB118details
A100 40GB40 GBFitsF168.0 GB212details
A100 80GB80 GBFitsF168.0 GB279details
H100 PCIe80 GBFitsF168.0 GB273details
H100 SXM80 GBFitsF168.0 GB458details
H200141 GBFitsF168.0 GB656details
RX 76008 GBFitsQ8_04.8 GB74details
RX 7700 XT12 GBFitsF168.0 GB59details
RX 7800 XT16 GBFitsF168.0 GB85details
RX 7900 GRE16 GBFitsF168.0 GB79details
RX 7900 XT20 GBFitsF168.0 GB109details
RX 907016 GBFitsF168.0 GB87details
RX 9070 XT16 GBFitsF168.0 GB87details
Ryzen AI Max+ 395 128GB128 GBFitsF168.0 GB35details
Arc A770 16GB16 GBFitsF168.0 GB77details
Mac M1 16GB16 GBFitsF168.0 GB9.3details
Mac M2 24GB24 GBFitsF168.0 GB14details
Mac M3 24GB24 GBFitsF168.0 GB14details
Mac M4 16GB16 GBFitsF168.0 GB16details
Mac M4 32GB32 GBFitsF168.0 GB16details
Mac M1 Pro 32GB32 GBFitsF168.0 GB27details
Mac M3 Pro 36GB36 GBFitsF168.0 GB20details
Mac M4 Pro 24GB24 GBFitsF168.0 GB37details
Mac M4 Pro 64GB64 GBFitsF168.0 GB37details
Mac M1 Max 64GB64 GBFitsF168.0 GB55details
Mac M2 Max 96GB96 GBFitsF168.0 GB55details
Mac M3 Max 128GB128 GBFitsF168.0 GB55details
Mac M4 Max 64GB64 GBFitsF168.0 GB75details
Mac M2 Ultra 192GB192 GBFitsF168.0 GB109details
Mac M3 Ultra 256GB256 GBFitsF168.0 GB112details

Questions

How much VRAM does granite 4.1 3b llmon preview need?

About 3.3 GB at 4-bit (Q4_K_M) with an 8K context: 2.1 GB of weights (real GGUF file size), 0.6 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 4.8 GB.

What is the smallest GPU that runs granite 4.1 3b llmon preview?

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 granite 4.1 3b llmon preview 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.

granite 4.2 3b

3.7B · 3.3 GB at 4-bit

granite 4.1 3b

3.4B · 3.2 GB at 4-bit

granite 4.0 micro

3.4B · 3.2 GB at 4-bit

Qwen2.5 VL 3B Instruct

3.8B · 2.8 GB at 4-bit

Raon 1 nano

3.5B · 3.0 GB at 4-bit

Phi 3 mini 4k instruct

3.8B · 5.7 GB at 4-bit

Model facts

Hugging Face repoibm-granite/granite-4.1-3b-llmon-preview
Parameters3,659,737,600 (3.7B)
Layers / hidden size40 / 2560
AttentionGQA; 40 heads, 8 KV heads, head dim 64
KV cache per token (FP16)80.0 KiB
Max context131,072 tokens
Licenseapache-2.0
Released on HFOct 1, 2026
Downloads (30d) / likes2,413 / 18

Sources: Hugging Face model API · config.json (ibm-granite/granite-4.1-3b-llmon-preview) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.