SmolCpp MoE 360M — Hardware Requirements & GPU Compatibility
ChatCode
SmolCpp MoE 360M is a 360M-parameter open language model from Edgeai-gio. At Q4_K_M it needs about 0.24 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 · 32K)Best on 24 GB (Q4_K_M · ~2730 tok/s · 32K)Macs from 8 GB (Q4_K_M · ~299 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run SmolCpp MoE 360M
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve Edgeai-gio/SmolCpp-MoE-360M --max-model-len 16384vLLM repo with its own quantisation
Download: Weights on Hugging Face
Other ways
each opens step-by-stepRun in cloud
Fits on RTX 3060 12GB (10 GB headroom) · Q4_K_M
- Generation speed
- ~975 tok/s
- generation speed
- Cost per 1M output tokens
- $0.022
- per 1M output tokens
How Much VRAM Does SmolCpp MoE 360M Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.2 GB | 2.5 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 32K context · ~3854 tok/s | |
| 0.2 GB | 2.5 GB | 3-bit medium — 3-bit medium quantization | Fits: 32K context · ~3448 tok/s | |
| 0.2 GB | 2.6 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 32K context · ~2730 tok/s | |
| 0.3 GB | 2.6 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 32K context · ~2340 tok/s | |
| 0.3 GB | 2.6 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 32K context · ~1986 tok/s | |
| 0.4 GB | 2.7 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 32K context · ~1638 tok/s | |
| 0.8 GB | 3.1 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 32K context · ~829 tok/s |
Selected Q4_K_M
4-bit medium quantization1.3GBat 16K context
- Weights
- 0.1 GB
- KV cache
- 0.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 22.7 GB free · ~2730 tok/s
≥50 tok/s needs ≥18 GB/s · RTX 5090 ~4853, RTX 3090 Ti ~2730
Which Hardware Can Run SmolCpp MoE 360M?
Q4_K_M · 0.2 GBSmolCpp MoE 360M (Q4_K_M) requires 0.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ 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 SmolCpp MoE 360M, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 6.7 GB free · up to 32K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~2059Simulate output on RTX 3080RTX 2080 Ti~1668Simulate output on RTX 2080 TiRTX 3070 Ti~1648Simulate output on RTX 3070 TiM3*8 GB~299Simulate output on iMac 24" M3 (8 GB)Orin Nano~276Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~199Simulate output on iPhone 1720 GB
fits · 18.7 GB free · up to 32K ctx
1 GPU
RX 7900 XT~2000Simulate output on RX 7900 XT32–40 GB
fits · 30.7 GB free · up to 32K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~4853Simulate output on RTX 5090M5 Max36 GB~1342Simulate output on Mac Studio M5 Max (36 GB)M532 GB~448Simulate output on MacBook Air 13" M5 (32 GB)Orin~555Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~366Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
5| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Granite 4.0 350MIBM | — | 1 GB | ~1150 | same size |
| SmolLM 360MHugging Face | — | 0.6 GB | ~1092 | same size |
| LFM2.5 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
| ERNIE 4.5 0.3B PTBaidu | — | 0.7 GB | ~1213 | same size |
| LFM2 350MLiquid AI | — | 0.8 GB | ~1130 | same size |
Frequently Asked Questions
How much VRAM does SmolCpp MoE 360M need?
SmolCpp MoE 360M requires 0.2 GB of VRAM at Q4_K_M, or 0.8 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 360M × 4.8 bits ÷ 8 = 0.2 GB
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 SmolCpp MoE 360M?
For SmolCpp MoE 360M, Q4_K_M (0.2 GB) offers the best balance of quality and VRAM usage. Q5_K_M (0.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.2 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run SmolCpp MoE 360M on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run SmolCpp MoE 360M. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.2 GB of usable unified memory (RAM minus macOS overhead).
Can I run SmolCpp MoE 360M locally?
Yes — SmolCpp MoE 360M can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is SmolCpp MoE 360M?
At Q4_K_M, SmolCpp MoE 360M can reach ~20000 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~2730 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 B200 → 8000 ÷ 0.24 × 0.65 = ~21667 tok/s
Estimated speed at Q4_K_M (0.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of SmolCpp MoE 360M?
At Q4_K_M, the download is about 0.22 GB. The full-precision BF16 version is 0.72 GB. The smallest option (Q2_K) is 0.15 GB.
Which GPUs can run SmolCpp MoE 360M?
52 consumer GPUs can run SmolCpp MoE 360M 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 SmolCpp MoE 360M?
163 devices with unified memory can run SmolCpp MoE 360M at Q4_K_M (0.2 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.