RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run granite 4.1 3b locally?

granite 4.1 3b by ibm-granite has 3.4B parameters. At 4-bit with an 8K context it needs about 3.2 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.

3.2 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 ibm-granite/granite-4.1-3b-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) •usable2.5 GB2.8 GB4.7 GB12.2 GB
4-bit (Q4_K_M) •good, most popular2.9 GB3.2 GB5.0 GB12.5 GB
5-bit (Q5_K_M) •very good3.2 GB3.5 GB5.4 GB12.9 GB
6-bit (Q6_K) •near lossless3.5 GB3.8 GB5.7 GB13.2 GB
8-bit (Q8_0) •effectively lossless4.3 GB4.6 GB6.5 GB14.0 GB
16-bit (F16/BF16) •original weights7.2 GB7.6 GB9.4 GB16.9 GB

Which GPUs and Macs run granite 4.1 3b?

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 GBFitsF167.6 GB53details
RTX 4060 Ti 16GB16 GBFitsF167.6 GB42details
RTX 309024 GBFitsF167.6 GB137details
RTX 409024 GBFitsF167.6 GB148details
RTX 509032 GBFitsF167.6 GB263details
RTX 5070 Ti16 GBFitsF167.6 GB132details
RX 7900 XTX24 GBFitsF167.6 GB141details
Mac M4 Pro 48GB48 GBFitsF167.6 GB40details
Mac M4 Max 128GB128 GBFitsF167.6 GB80details
Mac M3 Ultra 512GB512 GBFitsF167.6 GB120details
DGX Spark128 GBFitsF167.6 GB40details
RTX 407012 GBFitsF167.6 GB74details
RTX 5060 Ti 16GB16 GBFitsF167.6 GB66details
RTX 40608 GBFitsQ8_04.6 GB75details
Arc B58012 GBFitsF167.6 GB67details
RTX 3060 Ti8 GBFitsQ8_04.6 GB124details
RTX 30708 GBFitsQ8_04.6 GB124details
RTX 3080 10GB10 GBFitsF167.6 GB112details
RTX 3080 12GB12 GBFitsF167.6 GB134details
RTX 3090 Ti24 GBFitsF167.6 GB148details
RTX 4060 Ti 8GB8 GBFitsQ8_04.6 GB80details
RTX 4070 Super12 GBFitsF167.6 GB74details
RTX 4070 Ti12 GBFitsF167.6 GB74details
RTX 4070 Ti Super16 GBFitsF167.6 GB99details
RTX 408016 GBFitsF167.6 GB105details
RTX 4080 Super16 GBFitsF167.6 GB108details
RTX 50608 GBFitsQ8_04.6 GB124details
RTX 507012 GBFitsF167.6 GB99details
RTX 508016 GBFitsF167.6 GB141details
2x RTX 30902×24 GBFitsF167.6 GB137details
2x RTX 40902×24 GBFitsF167.6 GB148details
2x RTX 50902×32 GBFitsF167.6 GB263details
RTX A600048 GBFitsF167.6 GB113details
RTX 6000 Ada48 GBFitsF167.6 GB141details
RTX PRO 6000 Blackwell96 GBFitsF167.6 GB263details
L40S48 GBFitsF167.6 GB127details
A100 40GB40 GBFitsF167.6 GB228details
A100 80GB80 GBFitsF167.6 GB299details
H100 PCIe80 GBFitsF167.6 GB294details
H100 SXM80 GBFitsF167.6 GB492details
H200141 GBFitsF167.6 GB705details
RX 76008 GBFitsQ8_04.6 GB80details
RX 7700 XT12 GBFitsF167.6 GB63details
RX 7800 XT16 GBFitsF167.6 GB92details
RX 7900 GRE16 GBFitsF167.6 GB85details
RX 7900 XT20 GBFitsF167.6 GB117details
RX 907016 GBFitsF167.6 GB94details
RX 9070 XT16 GBFitsF167.6 GB94details
Ryzen AI Max+ 395 128GB128 GBFitsF167.6 GB38details
Arc A770 16GB16 GBFitsF167.6 GB82details
Mac M1 16GB16 GBFitsF167.6 GB10.0details
Mac M2 24GB24 GBFitsF167.6 GB15details
Mac M3 24GB24 GBFitsF167.6 GB15details
Mac M4 16GB16 GBFitsF167.6 GB18details
Mac M4 32GB32 GBFitsF167.6 GB18details
Mac M1 Pro 32GB32 GBFitsF167.6 GB29details
Mac M3 Pro 36GB36 GBFitsF167.6 GB22details
Mac M4 Pro 24GB24 GBFitsF167.6 GB40details
Mac M4 Pro 64GB64 GBFitsF167.6 GB40details
Mac M1 Max 64GB64 GBFitsF167.6 GB59details
Mac M2 Max 96GB96 GBFitsF167.6 GB59details
Mac M3 Max 128GB128 GBFitsF167.6 GB59details
Mac M4 Max 64GB64 GBFitsF167.6 GB80details
Mac M2 Ultra 192GB192 GBFitsF167.6 GB117details
Mac M3 Ultra 256GB256 GBFitsF167.6 GB120details

Questions

How much VRAM does granite 4.1 3b need?

About 3.2 GB at 4-bit (Q4_K_M) with an 8K context: 2.0 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 4.6 GB.

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

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 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 4.0 micro

3.4B · 3.2 GB at 4-bit

granite 3.1 3b a800m instruct

3.3B · 2.9 GB at 4-bit

granite 4.2 3b

3.7B · 3.3 GB at 4-bit

DeepSeek OCR 2

3.4B · 2.9 GB at 4-bit

deepseek vl2 tiny

3.4B · 3.0 GB at 4-bit

Unlimited OCR 2

3.3B · 3.0 GB at 4-bit

Model facts

Hugging Face repoibm-granite/granite-4.1-3b
Parameters3,402,836,480 (3.4B)
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 HFApr 6, 2026
Downloads (30d) / likes430,028 / 118

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