Snehanjan·Llama 3

Llama 3.2 3B Spider Text To Sql — Hardware Requirements & GPU Compatibility

Chat

Llama 3.2 3B Spider Text To Sql is a 3B-parameter open language model from Snehanjan in the Llama 3 family. At Q4_K_M it needs about 1.98 GB of VRAM — see which GPUs and Macs can run it below.

42 downloads 2 likes

Specifications

Publisher
Snehanjan
Family
Llama 3
Parameters
3B
Release Date
2026-06-02
License
llama3.2

Get Started

How Much VRAM Does Llama 3.2 3B Spider Text To Sql Need?

Select a quantization to see compatible GPUs below.

QuantizationBitsVRAM
Q2_Kest.3.401.4 GB
Q3_K_Mest.3.901.6 GB
Q4_K_Mest.4.802.0 GB
Q5_K_Mest.5.702.4 GB
Q6_Kest.6.602.7 GB
Q8_0est.8.003.3 GB
BF16est.16.006.6 GB

est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.

Which GPUs Can Run Llama 3.2 3B Spider Text To Sql?

Q4_K_M · 2.0 GB

Llama 3.2 3B Spider Text To Sql (Q4_K_M) requires 2.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. 50 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.

Runs great

Plenty of headroom
NVIDIA GeForce RTX 5090~588 tok/sNVIDIA GeForce RTX 3090 Ti~331 tok/sNVIDIA GeForce RTX 4090~331 tok/sNVIDIA GeForce RTX 5080~315 tok/sNVIDIA GeForce RTX 3090~307 tok/sNVIDIA GeForce RTX 3080 Ti~300 tok/sNVIDIA GeForce RTX 5070 Ti~294 tok/sNVIDIA GeForce RTX 5090 Laptop GPU~294 tok/sAMD Radeon RX 7900 XTX~267 tok/sNVIDIA GeForce RTX 3080~250 tok/sNVIDIA GeForce RTX 4080 SUPER~242 tok/sNVIDIA GeForce RTX 4080~235 tok/sAMD Radeon RX 7900 XT~222 tok/sNVIDIA GeForce RTX 4070 Ti SUPER~221 tok/sNVIDIA GeForce RTX 5070~221 tok/sNVIDIA TITAN RTX~221 tok/sNVIDIA GeForce RTX 2080 Ti~202 tok/sNVIDIA GeForce RTX 3070 Ti~200 tok/sNVIDIA GeForce RTX 4090 Laptop GPU~189 tok/sAMD Radeon RX 9070~178 tok/sAMD Radeon RX 9070 XT~178 tok/sAMD Radeon RX 7800 XT~173 tok/sNVIDIA GeForce RTX 4070~166 tok/sNVIDIA GeForce RTX 4070 SUPER~166 tok/sNVIDIA GeForce RTX 4070 Ti~166 tok/sAMD Radeon RX 7900 GRE~160 tok/sNVIDIA GeForce GTX 1080 Ti~159 tok/sNVIDIA GeForce RTX 3060 Ti~147 tok/sNVIDIA GeForce RTX 3070~147 tok/sNVIDIA GeForce RTX 5060~147 tok/sNVIDIA GeForce RTX 5060 Ti 16GB~147 tok/sNVIDIA GeForce RTX 5060 Ti 8GB~147 tok/sAMD Radeon RX 6800~142 tok/sAMD Radeon RX 6800 XT~142 tok/sAMD Radeon RX 6900 XT~142 tok/sIntel Arc A770 16GB~141 tok/sIntel Arc A750~129 tok/sAMD Radeon RX 7700 XT~120 tok/sNVIDIA GeForce RTX 3060 12GB~118 tok/sIntel Arc B580~115 tok/sAMD Radeon RX 6700 XT~107 tok/sIntel Arc B570~96 tok/sNVIDIA GeForce RTX 4060 Ti 16GB~95 tok/sNVIDIA GeForce RTX 4060 Ti 8GB~95 tok/sNVIDIA GeForce RTX 4060~89 tok/sAMD Radeon RX 9060 XT 16GB~89 tok/sAMD Radeon RX 7600~80 tok/sAMD Radeon RX 7600 XT~80 tok/sNVIDIA GeForce RTX 3060 8GB~79 tok/sNVIDIA GeForce RTX 3050 8GB~74 tok/s

Which Devices Can Run Llama 3.2 3B Spider Text To Sql?

Q4_K_M · 2.0 GB

59 devices with unified memory can run Llama 3.2 3B Spider Text To Sql, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.

Runs great

Plenty of headroom
NVIDIA DGX H100~8798 tok/sNVIDIA DGX A100 640GB~5355 tok/sMac Studio (M3 Ultra, 256GB)~290 tok/sMac Studio (M3 Ultra, 512GB)~290 tok/sMac Studio (M3 Ultra, 96GB)~290 tok/sMac Pro M2 Ultra (192 GB)~283 tok/sMac Studio M2 Ultra (192 GB)~283 tok/sMacBook Pro 16" M5 Max (128 GB)~217 tok/sMac Studio M4 Max (128 GB)~193 tok/sMac Studio M4 Max (64 GB)~193 tok/sMacBook Pro 16" M4 Max (48 GB)~193 tok/sMacBook Pro 16" M4 Max (64 GB)~193 tok/sMac Studio M4 Max (36 GB)~145 tok/sMacBook Pro 14" M4 Max (36 GB)~145 tok/sMacBook Pro 16" M3 Max (48 GB)~145 tok/sMacBook Pro 14-inch (M5 Pro)~109 tok/sMac Mini M4 Pro (24 GB)~97 tok/sMac Mini M4 Pro (48 GB)~97 tok/sMacBook Pro 14" M4 Pro (24 GB)~97 tok/sMacBook Pro 16" M4 Pro (24 GB)~97 tok/sASUS Ascent GX10~90 tok/sNVIDIA DGX Spark~90 tok/sNVIDIA Jetson AGX Thor Developer Kit~90 tok/sAsus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB)~84 tok/sBeelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB)~84 tok/sFramework Desktop (Ryzen AI Max+ 395, 128 GB)~84 tok/sGMKtec EVO-X2 (Ryzen AI Max+ 395, 128 GB)~84 tok/sHP Z2 Mini G1a (Ryzen AI Max+ PRO 395, 128 GB)~84 tok/sHP ZBook Ultra G1a 14 (Ryzen AI Max+ PRO 395, 128 GB)~84 tok/sMinisforum MS-S1 MAX (Ryzen AI Max+ 395, 128 GB)~84 tok/sSnapdragon X2 Elite Extreme Copilot+ PC~75 tok/sNVIDIA Jetson AGX Orin 32GB~67 tok/sNVIDIA Jetson AGX Orin 64GB~67 tok/sMacBook Pro 14-inch (M5)~54 tok/siPad Pro M5 13" (16 GB)~54 tok/sSnapdragon X Elite Copilot+ PC~44 tok/sMac Mini M4 (16 GB)~42 tok/sMac Mini M4 (32 GB)~42 tok/sMacBook Air 13" M4 (16 GB)~42 tok/sMacBook Air 13" M4 (24 GB)~42 tok/sMacBook Air 15" M4 (16 GB)~42 tok/sMacBook Air 15" M4 (24 GB)~42 tok/sMacBook Pro 14" M4 (16 GB)~42 tok/siPad Pro M4 13" (16 GB)~42 tok/sMacBook Air 13" M3 (16 GB)~36 tok/sMacBook Air 13" M3 (24 GB)~36 tok/sMacBook Air 13" M3 (8 GB)~36 tok/sIntel Core Ultra 9 288V (Lunar Lake) Laptop~35 tok/sNVIDIA Jetson Orin NX 16GB~34 tok/sNVIDIA Jetson Orin Nano 8GB (Super)~34 tok/sAMD Ryzen AI 9 HX 370 (Strix Point) Laptop~33 tok/sApple iPhone 17 Pro~27 tok/siPhone 17 Pro Max~27 tok/siPhone 17~24 tok/siPhone Air~24 tok/siPhone 15 ProiPhone 15 Pro MaxiPhone 16 ProiPhone 16 Pro Max

Related Models

Frequently Asked Questions

How much VRAM does Llama 3.2 3B Spider Text To Sql need?

Llama 3.2 3B Spider Text To Sql requires 2.0 GB of VRAM at Q4_K_M, or 6.6 GB at BF16.

VRAM = Weights + KV Cache + Overhead

Weights = 3B × 4.8 bits ÷ 8 = 1.8 GB

KV Cache + Overhead 0.2 GB (at 2K context + ~0.3 GB framework)

VRAM usage by quantization

2.0 GB

Learn more about VRAM estimation →

What's the best quantization for Llama 3.2 3B Spider Text To Sql?

For Llama 3.2 3B Spider Text To Sql, Q4_K_M (2.0 GB) offers the best balance of quality and VRAM usage. Q5_K_M (2.4 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.4 GB.

VRAM requirement by quantization

Q2_K
1.4 GB
Q4_K_M
2.0 GB
Q5_K_M
2.4 GB
Q6_K
2.7 GB
Q8_0
3.3 GB
BF16
6.6 GB

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run Llama 3.2 3B Spider Text To Sql on a Mac?

Llama 3.2 3B Spider Text To Sql requires at least 1.4 GB at Q2_K, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.

Can I run Llama 3.2 3B Spider Text To Sql locally?

Yes — Llama 3.2 3B Spider Text To Sql can run locally on consumer hardware. At Q4_K_M quantization it needs 2.0 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is Llama 3.2 3B Spider Text To Sql?

At Q4_K_M, Llama 3.2 3B Spider Text To Sql can reach ~2222 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~331 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = (bandwidth GB/s ÷ model GB) × efficiency

Example: NVIDIA B2008000 ÷ 2.0 × 0.65 = ~2626 tok/s

Estimated speed at Q4_K_M (2.0 GB)

~2626 tok/s
~331 tok/s
~2626 tok/s
~2222 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of Llama 3.2 3B Spider Text To Sql?

At Q4_K_M, the download is about 1.80 GB. The full-precision BF16 version is 6.00 GB. The smallest option (Q2_K) is 1.27 GB.

Which GPUs can run Llama 3.2 3B Spider Text To Sql?

50 consumer GPUs can run Llama 3.2 3B Spider Text To Sql at Q4_K_M (2.0 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 50 GPUs have plenty of headroom for comfortable inference.

Which devices can run Llama 3.2 3B Spider Text To Sql?

59 devices with unified memory can run Llama 3.2 3B Spider Text To Sql at Q4_K_M (2.0 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.