SmolLM3 3B ONNX — Hardware Requirements & GPU Compatibility
Chat
SmolLM3 3B ONNX is a 3B-parameter open language model from Hugging Face in the SmolLM family. It supports a context window of up to 65,536 tokens. At Q4_K_M it needs about 2.25 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 · 64K)Best on 24 GB (Q4_K_M · ~291 tok/s · 64K)Macs from 8 GB (Q4_K_M · ~32 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run SmolLM3 3B ONNX
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
vllm serve HuggingFaceTB/SmolLM3-3B-ONNX --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 (8 GB headroom) · Q4_K_M
- Generation speed
- ~104 tok/s
- generation speed
- Cost per 1M output tokens
- $0.22
- per 1M output tokens
How Much VRAM Does SmolLM3 3B ONNX Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.7 GB | 4 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 64K context · ~379 tok/s | |
| 1.9 GB | 4.2 GB | 3-bit medium — 3-bit medium quantization | Fits: 64K context · ~343 tok/s | |
| 2.3 GB | 4.5 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 64K context · ~291 tok/s | |
| 2.6 GB | 4.9 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 64K context · ~253 tok/s | |
| 2.9 GB | 5.2 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 64K context · ~224 tok/s | |
| 3.5 GB | 5.7 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 64K context · ~190 tok/s | |
| 6.5 GB | 8.7 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 64K context · ~102 tok/s |
Selected Q4_K_M
4-bit medium quantization3.3GBat 16K context
- Weights
- 1.8 GB
- KV cache
- 1.2 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 20.7 GB free · ~291 tok/s
≥50 tok/s needs ≥173 GB/s · RTX 5090 ~518, RTX 3090 Ti ~291
Which Hardware Can Run SmolLM3 3B ONNX?
Q4_K_M · 2.3 GBSmolLM3 3B ONNX (Q4_K_M) requires 2.3 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 66K context window can add up to 4.7 GB, bringing total usage to 6.9 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run SmolLM3 3B ONNX, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 4.7 GB free · up to 64K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~220Simulate output on RTX 3080RTX 2080 Ti~178Simulate output on RTX 2080 TiRTX 3070 Ti~176Simulate output on RTX 3070 TiM3*8 GB~32Simulate output on iMac 24" M3 (8 GB)Orin Nano~30Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~21Simulate output on iPhone 1720 GB
fits · 16.7 GB free · up to 64K ctx
1 GPU
RX 7900 XT~213Simulate output on RX 7900 XT32–40 GB
fits · 28.7 GB free · up to 64K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~518Simulate output on RTX 5090M5 Max36 GB~143Simulate output on Mac Studio M5 Max (36 GB)M532 GB~48Simulate output on MacBook Air 13" M5 (32 GB)Orin~59Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~39Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to SmolLM3 3B ONNX
SmolLM family · 15 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| D1 3BLiquid AI | — | 2.6 GB | ~285 | same size |
| Llama 3.2 3B InstructMeta | — | 4.1 GB | ~309 | same size |
| Phi 2Microsoft | — | 2.6 GB | ~248 | same size |
| AI21 Jamba Reasoning 3BAI21 Labs | — | 2.3 GB | ~291 | same size |
| Qwen2.5 3B InstructAlibaba | — | 2.8 GB | ~294 | same size |
| Quasar 3B A1B Previewsilx-ai | — | 8.4 GB | ~167 | same size |
Specifications
- Publisher
- Hugging Face
- Family
- SmolLM
- Parameters
- 3B
- Architecture
- SmolLM3ForCausalLM
- Context Length
- 65,536 tokens
- Vocabulary Size
- 128,256
- Release Date
- 2025-07-08
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does SmolLM3 3B ONNX need?
SmolLM3 3B ONNX requires 2.3 GB of VRAM at Q4_K_M, or 6.5 GB at BF16. Full 66K context adds up to 4.7 GB (6.9 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 3B × 4.8 bits ÷ 8 = 1.8 GB
KV Cache + Overhead ≈ 0.5 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.
KV Cache + Overhead ≈ 5.1 GB (at full 66K context)
VRAM usage by quantization
What's the best quantization for SmolLM3 3B ONNX?
For SmolLM3 3B ONNX, Q4_K_M (2.3 GB) offers the best balance of quality and VRAM usage. Q5_K_M (2.6 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.7 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run SmolLM3 3B ONNX on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run SmolLM3 3B ONNX. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.3 GB of usable unified memory (RAM minus macOS overhead).
Can I run SmolLM3 3B ONNX locally?
Yes — SmolLM3 3B ONNX can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is SmolLM3 3B ONNX?
At Q4_K_M, SmolLM3 3B ONNX can reach ~2133 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~291 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 ÷ 2.25 × 0.65 = ~2311 tok/s
Estimated speed at Q4_K_M (2.3 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of SmolLM3 3B ONNX?
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 SmolLM3 3B ONNX?
52 consumer GPUs can run SmolLM3 3B ONNX 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 SmolLM3 3B ONNX?
163 devices with unified memory can run SmolLM3 3B ONNX at Q4_K_M (2.3 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.