SmolLM2 360M Instruct — Hardware Requirements & GPU Compatibility
Chat
SmolLM2 360M Instruct is an instruction-tuned model from Hugging Face that occupies the sweet spot between the 135M and 1.7B entries in the SmolLM2 lineup. At 360 million parameters, it offers noticeably better coherence and instruction-following ability than the smallest variants while still running comfortably on virtually any modern GPU or even on CPU. This model is well suited for on-device assistants, embedded applications, and rapid prototyping where you need real conversational ability without dedicating significant hardware resources. It handles short-form generation, summarization, and basic reasoning tasks with reasonable quality.
189.5K downloads
181 likesSpecifications
- Publisher
- Hugging Face
- Parameters
- 360M
- Architecture
- LlamaForCausalLM
- Context Length
- 8,192 tokens
- Vocabulary Size
- 49,152
- Release Date
- 2025-09-22
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does SmolLM2 360M Instruct Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 1.1 GB | 1.4 GB | 0.72 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run SmolLM2 360M Instruct?
BF16 · 1.1 GBSmolLM2 360M Instruct (BF16) requires 1.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. Using the full 8K context window can add up to 0.3 GB, bringing total usage to 1.4 GB. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run SmolLM2 360M Instruct?
BF16 · 1.1 GB33 devices with unified memory can run SmolLM2 360M Instruct, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Derivatives (4)
Frequently Asked Questions
- How much VRAM does SmolLM2 360M Instruct need?
SmolLM2 360M Instruct requires 1.1 GB of VRAM at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 360M × 16 bits ÷ 8 = 0.7 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 0.7 GB (at full 8K context)
VRAM usage by quantization
BF161.1 GBBF16 + full context1.4 GB- Can I run SmolLM2 360M Instruct on a Mac?
SmolLM2 360M Instruct requires at least 1.1 GB at BF16, 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 SmolLM2 360M Instruct locally?
Yes — SmolLM2 360M Instruct can run locally on consumer hardware. At BF16 quantization it needs 1.1 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is SmolLM2 360M Instruct?
At BF16, SmolLM2 360M Instruct can reach ~2650 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~596 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 1.1 × 0.55 = ~2650 tok/s
Estimated speed at BF16 (1.1 GB)
AMD Instinct MI300X~2650 tok/sNVIDIA GeForce RTX 4090~596 tok/sNVIDIA H100 SXM~1981 tok/sAMD Instinct MI250X~1638 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of SmolLM2 360M Instruct?
At BF16, the download is about 0.72 GB.