SmolVLM2 256M Video Instruct — Hardware Requirements & GPU Compatibility
VisionSmolVLM2-256M-Video-Instruct is Hugging Face's multimodal model with 256 million parameters for analyzing video, images, and text to generate answers. It handles question answering over videos, captioning, comparing visuals, and transcribing image text, but it does not create images or video files. Fine-tuned from SmolVLM-256M-Instruct, the card notes that running video inference requires only about 1.38 GB of GPU memory. The page lists Video-MME at 33.7, MLVU at 40.6 and MVBench at 32.7, the lowest among the three SmolVLM2 sizes. It runs on nearly any hardware, including laptops, phone-class devices, and CPUs. Context length reaches 8,192 tokens. Released under the Apache 2.0 license, this allows unrestricted commercial and research use, with publication in February 2025. It is the smallest member of the SmolVLM2 family, which also features 500M and 2.2B versions; English is listed as its supported language on the card.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 8K)Best on 24 GB (Q4_K_M · ~1310 tok/s · 8K)Macs from 8 GB (Q4_K_M · ~143 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run SmolVLM2 256M Video Instruct
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/ggml-org/SmolVLM2-256M-Video-Instruct-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 23.4 GB free · 8K context
Download: SmolVLM2-256M-Video-Instruct-Q4_K_M.gguf0.13 GBggml-org
Ways to runOther ways
each opens step-by-stepRun in cloud
Your own private copy · Q4_K_M · 24 GB GPU
- The easiest way to try this model
- Running in a few minutes, easy setup
- Chat right here or connect any app
$5+ bonus for new users
Runs on RTX 3060 12GB (11 GB headroom) · Q4_K_M
- Generation speed
- ~468 tok/s
- generation speed
- Cost per 1M output tokens
- $0.037
- per 1M output tokens
How Much VRAM Does SmolVLM2 256M Video Instruct Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 0.5 GB | — | 3-bit small — 3-bit small quantization | Fits: 8K context · ~1424 tok/s | |
| 0.5 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 8K context · ~1424 tok/s | |
| 0.5 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 8K context · ~1394 tok/s | |
| 0.5 GB | — | 4-bit · legacy — 4-bit legacy quantization | Fits: 8K context · ~1365 tok/s | |
| 0.5 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 8K context · ~1310 tok/s | |
| 0.5 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 8K context · ~1236 tok/s | |
| 0.6 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 8K context · ~1170 tok/s | |
| 0.6 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 8K context · ~1092 tok/s |
Selected Q4_K_M
4-bit medium quantization0.6GBat 8K context
- Weights
- 0.2 GB
- KV cache
- 0.2 GB
- Overhead
- 0.3 GB
Which Hardware Can Run SmolVLM2 256M Video Instruct?
Q4_K_M · 0.5 GBSmolVLM2 256M Video Instruct (Q4_K_M) requires 0.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. Using the full 8K context window can add up to 0.1 GB, bringing total usage to 0.6 GB. 114 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 185 devices with unified memory can run SmolVLM2 256M Video Instruct, including Mac Studio M5 Ultra (256 GB), Mac Studio M5 Ultra (512 GB).
≤ 11 GB
≤ 11 GB tier →fits · 3.4 GB free · up to 8K ctx
68 GPUs · 1 Mac class · 3 device families
RTX 3080~988Simulate output on RTX 3080RTX 2080 Ti~801Simulate output on RTX 2080 TiRTX 3070 Ti~791Simulate output on RTX 3070 TiM3*8 GB~143Simulate output on iMac 24" M3 (8 GB)Orin Nano~133Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~96Simulate output on iPhone 1720 GB
fits · 19.4 GB free · up to 8K ctx
1 GPU
RX 7900 XT~960Simulate output on RX 7900 XT32–40 GB
fits · 31.4 GB free · up to 8K ctx
1 GPU · 3 Mac classes · 9 device families
RTX 5090~2330Simulate output on RTX 5090M5 Max*36 GB~644Simulate output on Mac Studio M5 Max (36 GB)M632 GB~238Simulate output on iMac 24" M6 (32 GB)RTX Spark N1X~391Simulate output on NVIDIA RTX Spark N1X Laptop (32 GB)Orin~266Simulate output on NVIDIA Jetson AGX Orin 32GB
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
3| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Gemma 3 270M ITGoogle | — | 1.2 GB | ~3640 | same size |
| Functiongemma 270M ITGoogle | — | 1.6 GB | ~1111 | same size |
| Tiny MixtralTitanML | — | 0.5 GB | ~2329 | same size |
Specifications
- Publisher
- Hugging Face
- Parameters
- 256M
- Architecture
- SmolVLMForConditionalGeneration
- Context Length
- 8,192 tokens
- Vocabulary Size
- 49,280
- Release Date
- 2025-02-11
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does SmolVLM2 256M Video Instruct need?
SmolVLM2 256M Video Instruct requires 0.5 GB of VRAM at Q4_K_M, or 0.9 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 256M × 4.8 bits ÷ 8 = 0.2 GB
KV Cache + Overhead ≈ 0.3 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 ≈ 0.4 GB (at full 8K context)
VRAM usage by quantization
What's the best quantization for SmolVLM2 256M Video Instruct?
For SmolVLM2 256M Video Instruct, Q4_K_M (0.5 GB) offers the best balance of quality and VRAM usage. Q5_0 (0.5 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XS at 0.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run SmolVLM2 256M Video Instruct on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 158 other Macs can run SmolVLM2 256M Video Instruct. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 0.5 GB of usable unified memory (RAM minus macOS overhead).
Can I run SmolVLM2 256M Video Instruct locally?
Yes — SmolVLM2 256M Video Instruct can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is SmolVLM2 256M Video Instruct?
At Q4_K_M, SmolVLM2 256M Video Instruct can reach ~9600 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~1310 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 ÷ 0.5 × 0.65 = ~10400 tok/s
Estimated speed at Q4_K_M (0.5 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of SmolVLM2 256M Video Instruct?
At Q4_K_M, the download is about 0.15 GB. The full-precision BF16 version is 0.51 GB. The smallest option (IQ2_XS) is 0.08 GB.
Which GPUs can run SmolVLM2 256M Video Instruct?
114 consumer GPUs can run SmolVLM2 256M Video Instruct at Q4_K_M. Top options include AMD Radeon RX 5600 XT, AMD Radeon RX 5700 XT, AMD Radeon RX 580 8GB. 114 GPUs have plenty of headroom for comfortable inference.
Which devices can run SmolVLM2 256M Video Instruct?
185 devices with unified memory can run SmolVLM2 256M Video Instruct at Q4_K_M (0.5 GB), including AMD Ryzen 7 8845HS (Hawk Point) Laptop, AMD Ryzen AI 7 350 (Krackan Point) Laptop, AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.