RunsOnMy ModelsGPUs & MacsAPI pricesVRAM calculatorNew

Can I run SmolLM2 360M Instruct locally?

SmolLM2 360M Instruct by HuggingFaceTB has 362M parameters. At 4-bit with an 8K context it needs about 1.1 GB of GPU memory. It fits on 65 of 65 GPUs and Macs we track.

1.1 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest Mac that fits
8Kmax 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/SmolLM2-360M-Instruct-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx
2-bit (Q2_K) •rough0.9 GB1.0 GB
3-bit (Q3_K_M) •usable0.9 GB1.0 GB
4-bit (Q4_K_M) •good, most popular0.9 GB1.1 GB
5-bit (Q5_K_M) •very good0.9 GB1.1 GB
6-bit (Q6_K) •near lossless1.0 GB1.1 GB
8-bit (Q8_0) •effectively lossless1.0 GB1.2 GB
16-bit (F16/BF16) •original weights1.3 GB1.5 GB

Which GPUs and Macs run SmolLM2 360M Instruct?

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 GBFitsF161.5 GB496details
RTX 4060 Ti 16GB16 GBFitsF161.5 GB397details
RTX 309024 GBFitsF161.5 GB1290details
RTX 409024 GBFitsF161.5 GB1389details
RTX 509032 GBFitsF161.5 GB2470details
RTX 5070 Ti16 GBFitsF161.5 GB1235details
RX 7900 XTX24 GBFitsF161.5 GB1323details
Mac M4 Pro 48GB48 GBFitsF161.5 GB376details
Mac M4 Max 128GB128 GBFitsF161.5 GB753details
Mac M3 Ultra 512GB512 GBFitsF161.5 GB1129details
DGX Spark128 GBFitsF161.5 GB376details
RTX 407012 GBFitsF161.5 GB695details
RTX 5060 Ti 16GB16 GBFitsF161.5 GB617details
RTX 40608 GBFitsF161.5 GB375details
Arc B58012 GBFitsF161.5 GB628details
RTX 3060 Ti8 GBFitsF161.5 GB617details
RTX 30708 GBFitsF161.5 GB617details
RTX 3080 10GB10 GBFitsF161.5 GB1047details
RTX 3080 12GB12 GBFitsF161.5 GB1257details
RTX 3090 Ti24 GBFitsF161.5 GB1389details
RTX 4060 Ti 8GB8 GBFitsF161.5 GB397details
RTX 4070 Super12 GBFitsF161.5 GB695details
RTX 4070 Ti12 GBFitsF161.5 GB695details
RTX 4070 Ti Super16 GBFitsF161.5 GB926details
RTX 408016 GBFitsF161.5 GB988details
RTX 4080 Super16 GBFitsF161.5 GB1014details
RTX 50608 GBFitsF161.5 GB617details
RTX 507012 GBFitsF161.5 GB926details
RTX 508016 GBFitsF161.5 GB1323details
2x RTX 30902×24 GBFitsF161.5 GB1290details
2x RTX 40902×24 GBFitsF161.5 GB1389details
2x RTX 50902×32 GBFitsF161.5 GB2470details
RTX A600048 GBFitsF161.5 GB1059details
RTX 6000 Ada48 GBFitsF161.5 GB1323details
RTX PRO 6000 Blackwell96 GBFitsF161.5 GB2470details
L40S48 GBFitsF161.5 GB1191details
A100 40GB40 GBFitsF161.5 GB2143details
A100 80GB80 GBFitsF161.5 GB2810details
H100 PCIe80 GBFitsF161.5 GB2757details
H100 SXM80 GBFitsF161.5 GB4617details
H200141 GBFitsF161.5 GB6616details
RX 76008 GBFitsF161.5 GB397details
RX 7700 XT12 GBFitsF161.5 GB595details
RX 7800 XT16 GBFitsF161.5 GB860details
RX 7900 GRE16 GBFitsF161.5 GB794details
RX 7900 XT20 GBFitsF161.5 GB1103details
RX 907016 GBFitsF161.5 GB882details
RX 9070 XT16 GBFitsF161.5 GB882details
Ryzen AI Max+ 395 128GB128 GBFitsF161.5 GB353details
Arc A770 16GB16 GBFitsF161.5 GB772details
Mac M1 16GB16 GBFitsF161.5 GB94details
Mac M2 24GB24 GBFitsF161.5 GB138details
Mac M3 24GB24 GBFitsF161.5 GB138details
Mac M4 16GB16 GBFitsF161.5 GB165details
Mac M4 32GB32 GBFitsF161.5 GB165details
Mac M1 Pro 32GB32 GBFitsF161.5 GB276details
Mac M3 Pro 36GB36 GBFitsF161.5 GB207details
Mac M4 Pro 24GB24 GBFitsF161.5 GB376details
Mac M4 Pro 64GB64 GBFitsF161.5 GB376details
Mac M1 Max 64GB64 GBFitsF161.5 GB551details
Mac M2 Max 96GB96 GBFitsF161.5 GB551details
Mac M3 Max 128GB128 GBFitsF161.5 GB551details
Mac M4 Max 64GB64 GBFitsF161.5 GB753details
Mac M2 Ultra 192GB192 GBFitsF161.5 GB1103details
Mac M3 Ultra 256GB256 GBFitsF161.5 GB1129details

Questions

How much VRAM does SmolLM2 360M Instruct need?

About 1.1 GB at 4-bit (Q4_K_M) with an 8K context: 0.3 GB of weights (real GGUF file size), 0.3 GB of KV cache and 0.5 GB of runtime overhead. At 8-bit it needs about 1.2 GB.

What is the smallest GPU that runs SmolLM2 360M Instruct?

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 SmolLM2 360M Instruct 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.

Model facts

Hugging Face repoHuggingFaceTB/SmolLM2-360M-Instruct
Parameters361,821,120 (362M)
Layers / hidden size32 / 960
AttentionGQA; 15 heads, 5 KV heads, head dim 64
KV cache per token (FP16)40.0 KiB
Max context8,192 tokens
Licenseapache-2.0
Released on HFOct 31, 2024
Downloads (30d) / likes227,479 / 248

Sources: Hugging Face model API · config.json (HuggingFaceTB/SmolLM2-360M-Instruct) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.