RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run Llama 3.2 1B locally?

Llama 3.2 1B by meta-llama has 1.2B parameters. At 4-bit with an 8K context it needs about 1.8 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.

1.8 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 QuantFactory/Dolphin3.0-Llama3.2-1B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K) •rough1.4 GB1.5 GB2.3 GB5.3 GB
3-bit (Q3_K_M) •usable1.5 GB1.6 GB2.4 GB5.4 GB
4-bit (Q4_K_M) •good, most popular1.7 GB1.8 GB2.5 GB5.5 GB
5-bit (Q5_K_M) •very good1.8 GB1.9 GB2.7 GB5.7 GB
6-bit (Q6_K) •near lossless1.9 GB2.0 GB2.8 GB5.8 GB
8-bit (Q8_0) •effectively lossless2.3 GB2.4 GB3.1 GB6.1 GB
16-bit (F16/BF16)original weights3.1 GB3.2 GB3.9 GB6.9 GB

Which GPUs and Macs run Llama 3.2 1B?

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 GBFitsF163.2 GB146details
RTX 4060 Ti 16GB16 GBFitsF163.2 GB117details
RTX 309024 GBFitsF163.2 GB379details
RTX 409024 GBFitsF163.2 GB408details
RTX 509032 GBFitsF163.2 GB725details
RTX 5070 Ti16 GBFitsF163.2 GB363details
RX 7900 XTX24 GBFitsF163.2 GB388details
Mac M4 Pro 48GB48 GBFitsF163.2 GB110details
Mac M4 Max 128GB128 GBFitsF163.2 GB221details
Mac M3 Ultra 512GB512 GBFitsF163.2 GB331details
DGX Spark128 GBFitsF163.2 GB110details
RTX 407012 GBFitsF163.2 GB204details
RTX 5060 Ti 16GB16 GBFitsF163.2 GB181details
RTX 40608 GBFitsF163.2 GB110details
Arc B58012 GBFitsF163.2 GB184details
RTX 3060 Ti8 GBFitsF163.2 GB181details
RTX 30708 GBFitsF163.2 GB181details
RTX 3080 10GB10 GBFitsF163.2 GB307details
RTX 3080 12GB12 GBFitsF163.2 GB369details
RTX 3090 Ti24 GBFitsF163.2 GB408details
RTX 4060 Ti 8GB8 GBFitsF163.2 GB117details
RTX 4070 Super12 GBFitsF163.2 GB204details
RTX 4070 Ti12 GBFitsF163.2 GB204details
RTX 4070 Ti Super16 GBFitsF163.2 GB272details
RTX 408016 GBFitsF163.2 GB290details
RTX 4080 Super16 GBFitsF163.2 GB298details
RTX 50608 GBFitsF163.2 GB181details
RTX 507012 GBFitsF163.2 GB272details
RTX 508016 GBFitsF163.2 GB388details
2x RTX 30902×24 GBFitsF163.2 GB379details
2x RTX 40902×24 GBFitsF163.2 GB408details
2x RTX 50902×32 GBFitsF163.2 GB725details
RTX A600048 GBFitsF163.2 GB311details
RTX 6000 Ada48 GBFitsF163.2 GB388details
RTX PRO 6000 Blackwell96 GBFitsF163.2 GB725details
L40S48 GBFitsF163.2 GB350details
A100 40GB40 GBFitsF163.2 GB629details
A100 80GB80 GBFitsF163.2 GB825details
H100 PCIe80 GBFitsF163.2 GB809details
H100 SXM80 GBFitsF163.2 GB1355details
H200141 GBFitsF163.2 GB1942details
RX 76008 GBFitsF163.2 GB117details
RX 7700 XT12 GBFitsF163.2 GB175details
RX 7800 XT16 GBFitsF163.2 GB252details
RX 7900 GRE16 GBFitsF163.2 GB233details
RX 7900 XT20 GBFitsF163.2 GB324details
RX 907016 GBFitsF163.2 GB259details
RX 9070 XT16 GBFitsF163.2 GB259details
Ryzen AI Max+ 395 128GB128 GBFitsF163.2 GB104details
Arc A770 16GB16 GBFitsF163.2 GB227details
Mac M1 16GB16 GBFitsF163.2 GB28details
Mac M2 24GB24 GBFitsF163.2 GB40details
Mac M3 24GB24 GBFitsF163.2 GB40details
Mac M4 16GB16 GBFitsF163.2 GB49details
Mac M4 32GB32 GBFitsF163.2 GB49details
Mac M1 Pro 32GB32 GBFitsF163.2 GB81details
Mac M3 Pro 36GB36 GBFitsF163.2 GB61details
Mac M4 Pro 24GB24 GBFitsF163.2 GB110details
Mac M4 Pro 64GB64 GBFitsF163.2 GB110details
Mac M1 Max 64GB64 GBFitsF163.2 GB162details
Mac M2 Max 96GB96 GBFitsF163.2 GB162details
Mac M3 Max 128GB128 GBFitsF163.2 GB162details
Mac M4 Max 64GB64 GBFitsF163.2 GB221details
Mac M2 Ultra 192GB192 GBFitsF163.2 GB324details
Mac M3 Ultra 256GB256 GBFitsF163.2 GB331details

Questions

How much VRAM does Llama 3.2 1B need?

About 1.8 GB at 4-bit (Q4_K_M) with an 8K context: 0.9 GB of weights (real GGUF file size), 0.3 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 2.4 GB.

What is the smallest GPU that runs Llama 3.2 1B?

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 Llama 3.2 1B 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.

Llama 3.2 1B Instruct

1.2B · 1.6 GB at 4-bit

Llama 3.2 3B Instruct

3.2B · 3.4 GB at 4-bit

Llama 3.2 3B

3.2B · 3.6 GB at 4-bit

LFM2.5 1.2B Instruct

1.2B · 1.3 GB at 4-bit

deepseek coder 1.3b instruct

1.3B · 2.8 GB at 4-bit

HunyuanOCR

1.1B · 2.0 GB at 4-bit

Model facts

Hugging Face repometa-llama/Llama-3.2-1B
Parameters1,235,814,400 (1.2B)
Layers / hidden size16 / 2048
AttentionGQA; 32 heads, 8 KV heads, head dim 64
KV cache per token (FP16)32.0 KiB
Max context131,072 tokens
Licensellama3.2 (gated: accept terms on Hugging Face)
Released on HFSep 18, 2024
Downloads (30d) / likes636,051 / 2,784

Sources: Hugging Face model API · config.json (unsloth/Llama-3.2-1B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.