RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can I run Groot 19B A3B 261023 locally?

Groot 19B A3B 261023 by GrootTeam has 19B parameters (3.0B active per token, 128 experts, 8 used per token). At 4-bit with an 8K context it needs about 12.7 GB of GPU memory. It fits on 59 of 65 GPUs and Macs we track. See all models that fit in 16 GB of VRAM.

12.7 GBQ4_K_M · 8K context
RTX 4060 Ti 16GBsmallest consumer GPU that fits
M2 24GBsmallest Mac that fits
256Kmax 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 bartowski/GrootTeam_Groot-19B-A3B-261023-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.

QuantQuality4K ctx8K ctx32K ctx128K ctx
2-bit (Q2_K) •rough8.3 GB8.4 GB8.8 GB10.7 GB
3-bit (Q3_K_M) •usable9.9 GB10.0 GB10.5 GB12.3 GB
4-bit (Q4_K_M) •good, most popular12.6 GB12.7 GB13.1 GB15.0 GB
5-bit (Q5_K_M) •very good14.7 GB14.7 GB15.2 GB17.1 GB
6-bit (Q6_K) •near lossless16.6 GB16.7 GB17.1 GB19.0 GB
8-bit (Q8_0) •effectively lossless19.8 GB19.9 GB20.3 GB22.2 GB
16-bit (F16/BF16) •original weights36.3 GB36.4 GB36.9 GB38.8 GB

Which GPUs and Macs run Groot 19B A3B 261023?

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 GBFitsQ3_K_M10.0 GB242details
RTX 4060 Ti 16GB16 GBFitsQ4_K_M12.7 GB149details
RTX 309024 GBFitsQ8_019.9 GB300details
RTX 409024 GBFitsQ8_019.9 GB323details
RTX 509032 GBFitsQ8_019.9 GB574details
RTX 5070 Ti16 GBFitsQ4_K_M12.7 GB465details
RX 7900 XTX24 GBFitsQ8_019.9 GB307details
Mac M4 Pro 48GB48 GBFitsQ8_019.9 GB87details
Mac M4 Max 128GB128 GBFitsF1636.4 GB93details
Mac M3 Ultra 512GB512 GBFitsF1636.4 GB140details
DGX Spark128 GBFitsF1636.4 GB47details
RTX 407012 GBFitsQ3_K_M10.0 GB339details
RTX 5060 Ti 16GB16 GBFitsQ4_K_M12.7 GB232details
RTX 40608 GBNo———details
Arc B58012 GBFitsQ3_K_M10.0 GB307details
RTX 3060 Ti8 GBNo———details
RTX 30708 GBNo———details
RTX 3080 10GB10 GBFitsQ2_K8.4 GB625details
RTX 3080 12GB12 GBFitsQ3_K_M10.0 GB614details
RTX 3090 Ti24 GBFitsQ8_019.9 GB323details
RTX 4060 Ti 8GB8 GBNo———details
RTX 4070 Super12 GBFitsQ3_K_M10.0 GB339details
RTX 4070 Ti12 GBFitsQ3_K_M10.0 GB339details
RTX 4070 Ti Super16 GBFitsQ4_K_M12.7 GB349details
RTX 408016 GBFitsQ4_K_M12.7 GB372details
RTX 4080 Super16 GBFitsQ4_K_M12.7 GB382details
RTX 50608 GBNo———details
RTX 507012 GBFitsQ3_K_M10.0 GB453details
RTX 508016 GBFitsQ4_K_M12.7 GB498details
2x RTX 30902×24 GBFitsF1636.4 GB160details
2x RTX 40902×24 GBFitsF1636.4 GB172details
2x RTX 50902×32 GBFitsF1636.4 GB305details
RTX A600048 GBFitsF1636.4 GB131details
RTX 6000 Ada48 GBFitsF1636.4 GB164details
RTX PRO 6000 Blackwell96 GBFitsF1636.4 GB305details
L40S48 GBFitsF1636.4 GB147details
A100 40GB40 GBFitsQ8_019.9 GB498details
A100 80GB80 GBFitsF1636.4 GB348details
H100 PCIe80 GBFitsF1636.4 GB341details
H100 SXM80 GBFitsF1636.4 GB571details
H200141 GBFitsF1636.4 GB818details
RX 76008 GBNo———details
RX 7700 XT12 GBFitsQ3_K_M10.0 GB291details
RX 7800 XT16 GBFitsQ4_K_M12.7 GB324details
RX 7900 GRE16 GBFitsQ4_K_M12.7 GB299details
RX 7900 XT20 GBFitsQ6_K16.7 GB309details
RX 907016 GBFitsQ4_K_M12.7 GB332details
RX 9070 XT16 GBFitsQ4_K_M12.7 GB332details
Ryzen AI Max+ 395 128GB128 GBFitsF1636.4 GB44details
Arc A770 16GB16 GBFitsQ4_K_M12.7 GB291details
Mac M1 16GB16 GBFitsQ2_K8.4 GB56details
Mac M2 24GB24 GBFitsQ4_K_M12.7 GB52details
Mac M3 24GB24 GBFitsQ4_K_M12.7 GB52details
Mac M4 16GB16 GBFitsQ2_K8.4 GB99details
Mac M4 32GB32 GBFitsQ6_K16.7 GB46details
Mac M1 Pro 32GB32 GBFitsQ6_K16.7 GB77details
Mac M3 Pro 36GB36 GBFitsQ8_019.9 GB48details
Mac M4 Pro 24GB24 GBFitsQ4_K_M12.7 GB142details
Mac M4 Pro 64GB64 GBFitsF1636.4 GB47details
Mac M1 Max 64GB64 GBFitsF1636.4 GB68details
Mac M2 Max 96GB96 GBFitsF1636.4 GB68details
Mac M3 Max 128GB128 GBFitsF1636.4 GB68details
Mac M4 Max 64GB64 GBFitsF1636.4 GB93details
Mac M2 Ultra 192GB192 GBFitsF1636.4 GB136details
Mac M3 Ultra 256GB256 GBFitsF1636.4 GB140details

Questions

How much VRAM does Groot 19B A3B 261023 need?

About 12.7 GB at 4-bit (Q4_K_M) with an 8K context: 11.6 GB of weights (real GGUF file size), 0.2 GB of KV cache and 0.9 GB of runtime overhead. At 8-bit it needs about 19.9 GB.

What is the smallest GPU that runs Groot 19B A3B 261023?

At Q4_K_M and 8K context, the smallest single consumer GPU in our table that fits it is the RTX 4060 Ti 16GB (16 GB).

Can I run Groot 19B A3B 261023 on a Mac?

Yes. At Q4_K_M with 8K context it fits on a Mac M2 24GB 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.

gpt oss 20b

21B · 11.8 GB at 4-bit

gpt oss safeguard 20b

22B · 11.8 GB at 4-bit

ERNIE 4.5 21B A3B Thinking

22B · 13.5 GB at 4-bit

ERNIE 4.5 21B A3B PT

22B · 13.5 GB at 4-bit

Kimi VL A3B Thinking 2506

16B · 10.8 GB at 4-bit

Kimi VL A3B Instruct

16B · 10.8 GB at 4-bit

Model facts

Hugging Face repoGrootTeam/Groot-19B-A3B-261023
Parameters19,432,294,256 (19B) · ~3.0B active (from the model name)
Layers / hidden size40 / 2048
AttentionGQA (hybrid: 30 linear/state layers); 16 heads, 2 KV heads, head dim 256
KV cache per token (FP16)20.0 KiB
Max context262,144 tokens
Licenseapache-2.0
Released on HFOct 23, 2026
Downloads (30d) / likes2,489 / 99

Sources: Hugging Face model API · config.json (GrootTeam/Groot-19B-A3B-261023) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.