RunsOnMy ModelsGPUs & MacsAPI pricesVRAM calculatorNew

Can I run d1 2B locally?

d1 2B by LiquidAI has 1.9B 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.

1.8 GBQ4_K_M · 8K context
RTX 4060smallest consumer GPU that fits
M1 16GBsmallest Mac that fits
125Kmax context
GQA + sliding windowattention type

Memory needed by quant and context length

GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from LiquidAI/d1-2B-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx
2-bit (Q2_K)rough1.4 GB1.4 GB1.7 GB
3-bit (Q3_K_M)usable1.6 GB1.6 GB1.9 GB
4-bit (Q4_K_M) •good, most popular1.7 GB1.8 GB2.0 GB
5-bit (Q5_K_M)very good2.0 GB2.0 GB2.3 GB
6-bit (Q6_K)near lossless2.2 GB2.2 GB2.5 GB
8-bit (Q8_0) •effectively lossless2.4 GB2.5 GB2.7 GB
16-bit (F16/BF16) •original weights7.2 GB7.2 GB7.5 GB

Which GPUs and Macs run d1 2B?

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 GBFitsF167.2 GB52details
RTX 4060 Ti 16GB16 GBFitsF167.2 GB42details
RTX 309024 GBFitsF167.2 GB135details
RTX 409024 GBFitsF167.2 GB145details
RTX 509032 GBFitsF167.2 GB258details
RTX 5070 Ti16 GBFitsF167.2 GB129details
RX 7900 XTX24 GBFitsF167.2 GB138details
Mac M4 Pro 48GB48 GBFitsF167.2 GB39details
Mac M4 Max 128GB128 GBFitsF167.2 GB79details
Mac M3 Ultra 512GB512 GBFitsF167.2 GB118details
DGX Spark128 GBFitsF167.2 GB39details
RTX 407012 GBFitsF167.2 GB73details
RTX 5060 Ti 16GB16 GBFitsF167.2 GB65details
RTX 40608 GBFitsQ8_02.5 GB147details
Arc B58012 GBFitsF167.2 GB66details
RTX 3060 Ti8 GBFitsQ8_02.5 GB243details
RTX 30708 GBFitsQ8_02.5 GB243details
RTX 3080 10GB10 GBFitsF167.2 GB110details
RTX 3080 12GB12 GBFitsF167.2 GB132details
RTX 3090 Ti24 GBFitsF167.2 GB145details
RTX 4060 Ti 8GB8 GBFitsQ8_02.5 GB156details
RTX 4070 Super12 GBFitsF167.2 GB73details
RTX 4070 Ti12 GBFitsF167.2 GB73details
RTX 4070 Ti Super16 GBFitsF167.2 GB97details
RTX 408016 GBFitsF167.2 GB103details
RTX 4080 Super16 GBFitsF167.2 GB106details
RTX 50608 GBFitsQ8_02.5 GB243details
RTX 507012 GBFitsF167.2 GB97details
RTX 508016 GBFitsF167.2 GB138details
2x RTX 30902×24 GBFitsF167.2 GB135details
2x RTX 40902×24 GBFitsF167.2 GB145details
2x RTX 50902×32 GBFitsF167.2 GB258details
RTX A600048 GBFitsF167.2 GB111details
RTX 6000 Ada48 GBFitsF167.2 GB138details
RTX PRO 6000 Blackwell96 GBFitsF167.2 GB258details
L40S48 GBFitsF167.2 GB125details
A100 40GB40 GBFitsF167.2 GB224details
A100 80GB80 GBFitsF167.2 GB294details
H100 PCIe80 GBFitsF167.2 GB288details
H100 SXM80 GBFitsF167.2 GB483details
H200141 GBFitsF167.2 GB692details
RX 76008 GBFitsQ8_02.5 GB156details
RX 7700 XT12 GBFitsF167.2 GB62details
RX 7800 XT16 GBFitsF167.2 GB90details
RX 7900 GRE16 GBFitsF167.2 GB83details
RX 7900 XT20 GBFitsF167.2 GB115details
RX 907016 GBFitsF167.2 GB92details
RX 9070 XT16 GBFitsF167.2 GB92details
Ryzen AI Max+ 395 128GB128 GBFitsF167.2 GB37details
Arc A770 16GB16 GBFitsF167.2 GB81details
Mac M1 16GB16 GBFitsF167.2 GB9.8details
Mac M2 24GB24 GBFitsF167.2 GB14details
Mac M3 24GB24 GBFitsF167.2 GB14details
Mac M4 16GB16 GBFitsF167.2 GB17details
Mac M4 32GB32 GBFitsF167.2 GB17details
Mac M1 Pro 32GB32 GBFitsF167.2 GB29details
Mac M3 Pro 36GB36 GBFitsF167.2 GB22details
Mac M4 Pro 24GB24 GBFitsF167.2 GB39details
Mac M4 Pro 64GB64 GBFitsF167.2 GB39details
Mac M1 Max 64GB64 GBFitsF167.2 GB58details
Mac M2 Max 96GB96 GBFitsF167.2 GB58details
Mac M3 Max 128GB128 GBFitsF167.2 GB58details
Mac M4 Max 64GB64 GBFitsF167.2 GB79details
Mac M2 Ultra 192GB192 GBFitsF167.2 GB115details
Mac M3 Ultra 256GB256 GBFitsF167.2 GB118details

Questions

How much VRAM does d1 2B need?

About 1.8 GB at 4-bit (Q4_K_M) with an 8K context: 1.0 GB of weights (real GGUF file size), 0.1 GB of KV cache and 0.6 GB of runtime overhead. At 8-bit it needs about 2.5 GB.

What is the smallest GPU that runs d1 2B?

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 d1 2B 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 repoLiquidAI/d1-2B
Parameters1,918,181,888 (1.9B)
Layers / hidden size24 / 2048
AttentionGQA + sliding window (hybrid: 13 linear/state layers); 16 heads, 4 KV heads, head dim 128; sliding window 256
KV cache per token (FP16)22.0 KiB (sliding layers capped at the window)
Max context128,000 tokens
Licenseother
Released on HFOct 24, 2026
Downloads (30d) / likes14,128 / 293

Sources: Hugging Face model API · config.json (LiquidAI/d1-2B) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.