SKT ST X 0 3B — Hardware Requirements & GPU Compatibility
Chat
SKT ST X 0 3B is a 3.4B-parameter open language model from sKT-Ai-Labs. It supports a context window of up to 2,048 tokens. At Q4_K_M it needs about 2.38 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 2K)Best on 24 GB (Q4_K_M · ~312 tok/s · 2K)Macs from 8 GB (Q4_K_M · ~44 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run SKT ST X 0 3B
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve sKT-Ai-Labs/SKT-ST-X-0-3B --max-model-len 2048vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (9 GB headroom) · Q4_K_M
- Generation speed
- ~147 tok/s
- generation speed
- Cost per 1M output tokens
- $0.16
- per 1M output tokens
How Much VRAM Does SKT ST X 0 3B Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.8 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 2K context · ~370 tok/s | |
| 2 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 2K context · ~346 tok/s | |
| 2.4 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 2K context · ~312 tok/s | |
| 2.8 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 2K context · ~284 tok/s | |
| 3.1 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 2K context · ~260 tok/s | |
| 3.7 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 2K context · ~230 tok/s | |
| 7.1 GB | — | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 2K context · ~139 tok/s |
Selected Q4_K_M
4-bit medium quantization2.4GBat 2K context
- Weights
- 2.0 GB
- KV cache
- 0.1 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 21.6 GB free · ~312 tok/s
≥50 tok/s needs ≥107 GB/s · RTX 5090 ~429, RTX 3090 Ti ~312
Which Hardware Can Run SKT ST X 0 3B?
Q4_K_M · 2.4 GBSKT ST X 0 3B (Q4_K_M) requires 2.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 4+ GB is recommended. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run SKT ST X 0 3B, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 5.6 GB free · up to 2K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~259Simulate output on RTX 3080RTX 2080 Ti~223Simulate output on RTX 2080 TiRTX 3070 Ti~221Simulate output on RTX 3070 TiM3*8 GB~44Simulate output on iMac 24" M3 (8 GB)Orin Nano~48Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~31Simulate output on iPhone 1720 GB
fits · 17.6 GB free · up to 2K ctx
1 GPU
RX 7900 XT~140Simulate output on RX 7900 XT32–40 GB
fits · 29.6 GB free · up to 2K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~429Simulate output on RTX 5090M5 Max36 GB~118Simulate output on Mac Studio M5 Max (36 GB)M532 GB~60Simulate output on MacBook Air 13" M5 (32 GB)Orin~91Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~58Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite 4.0 MicroIBM | — | 3.7 GB | ~261 | same size |
| Deepseek Vl2 TinyDeepSeek | — | 9.4 GB | ~152 | same size |
| Llama 3.2 3B InstructMeta | — | 4.1 GB | ~309 | same size |
| AI21 Jamba Reasoning 3BAI21 Labs | — | 2.3 GB | ~291 | same size |
| MiniMax M3 EAGLE3Inferact | — | 2.7 GB | ~284 | same size |
| Tiny Aya BaseCohere | — | 9.4 GB | ~89 | same size |
Specifications
- Publisher
- sKT-Ai-Labs
- Parameters
- 3.4B
- Active parameters
- 1.9B per token
- Architecture
- MixtralForCausalLM
- Context Length
- 2,048 tokens
- Vocabulary Size
- 32,000
- Release Date
- 2026-06-03
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does SKT ST X 0 3B need?
SKT ST X 0 3B requires 2.4 GB of VRAM at Q4_K_M, or 7.1 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 3.4B × 4.8 bits ÷ 8 = 2 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
VRAM usage by quantization
What's the best quantization for SKT ST X 0 3B?
For SKT ST X 0 3B, Q4_K_M (2.4 GB) offers the best balance of quality and VRAM usage. Q5_K_M (2.8 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.8 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run SKT ST X 0 3B on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run SKT ST X 0 3B. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.4 GB of usable unified memory (RAM minus macOS overhead).
Can I run SKT ST X 0 3B locally?
Yes — SKT ST X 0 3B can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is SKT ST X 0 3B?
At Q4_K_M, SKT ST X 0 3B can reach ~214 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~312 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 1.3 GB active ÷ (8000 × 0.65) = 0.252 ms, plus 22 layers × 0.055 ms = 1.210 ms, so 1000 ÷ 1.462 ms = ~684 tok/s
Estimated speed at Q4_K_M (2.4 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of SKT ST X 0 3B?
At Q4_K_M, the download is about 2.03 GB. The full-precision BF16 version is 6.77 GB. The smallest option (Q2_K) is 1.44 GB.
Which GPUs can run SKT ST X 0 3B?
52 consumer GPUs can run SKT ST X 0 3B at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run SKT ST X 0 3B?
163 devices with unified memory can run SKT ST X 0 3B at Q4_K_M (2.4 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (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.