RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run Llama 3.1 70B locally?

Llama 3.1 70B by meta-llama has 71B parameters. At 4-bit with an 8K context it needs about 42.7 GB of GPU memory. It fits on 25 of 65 GPUs and Macs we track. See all models that fit in 48 GB of VRAM.

42.7 GBQ4_K_M · 8K context
nonesmallest consumer GPU that fits
M4 Pro 64GBsmallest 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 ThomasBaruzier/Llama-3.3-70B-Instruct-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K) •rough26.5 GB27.7 GB35.2 GB65.2 GB
3-bit (Q3_K_M) •usable33.8 GB35.1 GB42.6 GB72.6 GB
4-bit (Q4_K_M) •good, most popular41.5 GB42.7 GB50.2 GB80.2 GB
5-bit (Q5_K_M) •very good48.4 GB49.7 GB57.2 GB87.2 GB
6-bit (Q6_K) •near lossless55.8 GB57.1 GB64.6 GB94.6 GB
8-bit (Q8_0) •effectively lossless71.7 GB73.0 GB80.5 GB110 GB
16-bit (F16/BF16) •original weights133 GB135 GB142 GB172 GB

Which GPUs and Macs run Llama 3.1 70B?

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 GBNo———details
RTX 309024 GBNo———details
RTX 409024 GBNo———details
RTX 509032 GBTightQ2_K27.7 GB68details
RTX 5070 Ti16 GBNo———details
RX 7900 XTX24 GBNo———details
Mac M4 Pro 48GB48 GBFitsQ2_K27.7 GB10details
Mac M4 Max 128GB128 GBFitsQ8_073.0 GB7.3details
Mac M3 Ultra 512GB512 GBFitsF16135 GB5.8details
DGX Spark128 GBFitsQ8_073.0 GB3.6details
RTX 407012 GBNo———details
RTX 5060 Ti 16GB16 GBNo———details
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 GBNo———details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBNo———details
RTX 4070 Ti12 GBNo———details
RTX 4070 Ti Super16 GBNo———details
RTX 408016 GBNo———details
RTX 4080 Super16 GBNo———details
RTX 50608 GBNo———details
RTX 507012 GBNo———details
RTX 508016 GBNo———details
2x RTX 30902×24 GBFitsQ3_K_M35.1 GB27details
2x RTX 40902×24 GBFitsQ3_K_M35.1 GB29details
2x RTX 50902×32 GBFitsQ5_K_M49.7 GB36details
RTX A600048 GBFitsQ3_K_M35.1 GB22details
RTX 6000 Ada48 GBFitsQ3_K_M35.1 GB28details
RTX PRO 6000 Blackwell96 GBFitsQ8_073.0 GB24details
L40S48 GBFitsQ3_K_M35.1 GB25details
A100 40GB40 GBFitsQ2_K27.7 GB59details
A100 80GB80 GBFitsQ6_K57.1 GB35details
H100 PCIe80 GBFitsQ6_K57.1 GB35details
H100 SXM80 GBFitsQ6_K57.1 GB58details
H200141 GBFitsQ8_073.0 GB64details
RX 76008 GBNo———details
RX 7700 XT12 GBNo———details
RX 7800 XT16 GBNo———details
RX 7900 GRE16 GBNo———details
RX 7900 XT20 GBNo———details
RX 907016 GBNo———details
RX 9070 XT16 GBNo———details
Ryzen AI Max+ 395 128GB128 GBFitsQ8_073.0 GB3.4details
Arc A770 16GB16 GBNo———details
Mac M1 16GB16 GBNo———details
Mac M2 24GB24 GBNo———details
Mac M3 24GB24 GBNo———details
Mac M4 16GB16 GBNo———details
Mac M4 32GB32 GBNo———details
Mac M1 Pro 32GB32 GBNo———details
Mac M3 Pro 36GB36 GBNo———details
Mac M4 Pro 24GB24 GBNo———details
Mac M4 Pro 64GB64 GBFitsQ4_K_M42.7 GB6.4details
Mac M1 Max 64GB64 GBFitsQ4_K_M42.7 GB9.4details
Mac M2 Max 96GB96 GBFitsQ6_K57.1 GB6.9details
Mac M3 Max 128GB128 GBFitsQ8_073.0 GB5.3details
Mac M4 Max 64GB64 GBFitsQ4_K_M42.7 GB13details
Mac M2 Ultra 192GB192 GBFitsQ8_073.0 GB11details
Mac M3 Ultra 256GB256 GBFitsF16135 GB5.8details

Questions

How much VRAM does Llama 3.1 70B need?

About 42.7 GB at 4-bit (Q4_K_M) with an 8K context: 39.6 GB of weights (real GGUF file size), 2.5 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 73.0 GB.

What is the smallest GPU that runs Llama 3.1 70B?

No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the Mac M4 Pro 64GB (64 GB).

Can I run Llama 3.1 70B on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M4 Pro 64GB 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 3.3 70B Instruct

71B · 43.3 GB at 4-bit

Llama 4 Scout 17B 16E Instruct

109B · 63.2 GB at 4-bit

Llama 4 Scout 17B 16E

109B · 64.1 GB at 4-bit

DeepSeek R1 Distill Llama 70B

71B · 43.3 GB at 4-bit

Kimi Dev 72B

73B · 44.6 GB at 4-bit

Hunyuan A13B Instruct

80B · 47.1 GB at 4-bit

Model facts

Hugging Face repometa-llama/Llama-3.1-70B
Parameters70,553,706,496 (71B)
Layers / hidden size80 / 8192
AttentionGQA; 64 heads, 8 KV heads, head dim 128
KV cache per token (FP16)320.0 KiB
Max context131,072 tokens
Licensellama3.1 (gated: accept terms on Hugging Face)
Released on HFJul 14, 2024
Downloads (30d) / likes20,283 / 454

Sources: Hugging Face model API · config.json (unsloth/Llama-3.1-70B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.