litert-community / SmolVLM2-500M

huggingface.co
Total runs: 251
24-hour runs: 0
7-day runs: 2
30-day runs: 2
Model's Last Updated: July 03 2026
image-text-to-text

Introduction of SmolVLM2-500M

Model Details of SmolVLM2-500M

SmolVLM2-500M — LiteRT-LM (on-device Vision-Language Model)

HuggingFaceTB/SmolVLM2-500M-Video-Instruct (image path) converted to the LiteRT-LM ( .litertlm ) format for on-device image+text inference with Google's LiteRT-LM runtime.

SmolVLM2-500M is a tiny vision-language model from Hugging Face: a SigLIP vision encoder + pixel-shuffle connector feeding a SmolLM2 (Llama-architecture) 360M decoder. At just 361 MB it is one of the smallest on-device VLMs — give it an image and a question, get a grounded answer, fully offline.

File SmolVLM2-500M.litertlm (~361 MB)
Vision SigLIP encoder (512×512, 1024 patches, no CLS) + pixel-shuffle ×4 + Linear connector, int8 → 64 image tokens
Decoder SmolLM2-360M (Llama, 960-dim, 32 layers, GQA 15/5), int4 weights (blockwise-32 + OCTAV); tied embedding INT8 (externalized)
Compute integer
Context (KV cache) 2048
Image input resized to 512×512 ((x−0.5)/0.5 normalization baked into the vision encoder)
Base model HuggingFaceTB/SmolVLM2-500M-Video-Instruct
Quality

Single-image VQA produces coherent, image-grounded answers (CPU-verified; the SigLIP vision tower converts bit-faithfully, float CPU-parity corr ≈ 1.0). It is a very small (500M) model — keep a sensible max_tokens and use sampling (e.g. top-p); at pure greedy it can be repetitive/verbose.

⚠️ Best for single-image VQA — one image per conversation

Ask about one image per chat (start a new conversation for a different image). Single-image VQA is the primary use case. (On the GPU backend, a second image in the same conversation may degrade — a GPU-delegate trait shared across fast_vlm models; CPU handles multi-image.)

Run on iPhone / macOS

Use the LiteRT-LM Swift runtime ( swift-litert-lm / the LiteRTDemo sample). Load SmolVLM2-500M.litertlm with the vision tower enabled (modalities Modality.textImage / [.vision] — vision-only bundle, no audio tower), attach a photo, ask a question.

Run on Android — Google AI Edge Gallery

Run this model with image input in the official Google AI Edge Gallery app — no custom app needed (the bundle carries the tokenizer, chat template, and image preprocessing config):

  1. Push the bundle onto the phone (or download it there directly from this repo): adb push SmolVLM2-500M.litertlm /sdcard/Download/
  2. Open the Gallery app, tap the + icon (bottom-right) and pick SmolVLM2-500M.litertlm in the file picker.
  3. In the Import Model dialog, check "Support image" (required for image input), set a sensible max tokens , pick GPU (fast) or CPU , then tap Import .
  4. Open the Ask Image task, choose the imported model, attach a photo, and ask.

Tip: ask about one image per conversation . It's a tiny 500M model — keep max-tokens modest so it doesn't ramble.

Conversion notes
  • LiteRT-LM fast_vlm bundle: VISION_ENCODER ( [1,512,512,3] → [1,1024,768] , SigLIP) + VISION_ADAPTER ( [1,1024,768] → [1,64,960] , pixel-shuffle ×4 + Linear) + single-token EMBEDDER + PREFILL_DECODE.
  • The vision encoder uses the static arange(1024) position-embedding path (the model's dynamic bucketize position logic is bypassed — numerically identical for a full 512×512 frame) and bakes the (x−0.5)/0.5 normalization + NCHW transpose into the graph.
  • Single-image, no high-res splitting → a fixed 64 soft tokens; SmolLM2 (Llama) decoder exported with externalized (tied) embedder.
License

Apache-2.0 (SmolVLM2 + SmolLM2). See the base model card .

Runs of litert-community SmolVLM2-500M on huggingface.co

251
Total runs
0
24-hour runs
1
3-day runs
2
7-day runs
2
30-day runs

More Information About SmolVLM2-500M huggingface.co Model

More SmolVLM2-500M license Visit here:

https://choosealicense.com/licenses/apache-2.0

SmolVLM2-500M huggingface.co

SmolVLM2-500M huggingface.co is an AI model on huggingface.co that provides SmolVLM2-500M's model effect (), which can be used instantly with this litert-community SmolVLM2-500M model. huggingface.co supports a free trial of the SmolVLM2-500M model, and also provides paid use of the SmolVLM2-500M. Support call SmolVLM2-500M model through api, including Node.js, Python, http.

litert-community SmolVLM2-500M online free

SmolVLM2-500M huggingface.co is an online trial and call api platform, which integrates SmolVLM2-500M's modeling effects, including api services, and provides a free online trial of SmolVLM2-500M, you can try SmolVLM2-500M online for free by clicking the link below.

litert-community SmolVLM2-500M online free url in huggingface.co:

https://huggingface.co/litert-community/SmolVLM2-500M

SmolVLM2-500M install

SmolVLM2-500M is an open source model from GitHub that offers a free installation service, and any user can find SmolVLM2-500M on GitHub to install. At the same time, huggingface.co provides the effect of SmolVLM2-500M install, users can directly use SmolVLM2-500M installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

SmolVLM2-500M install url in huggingface.co:

https://huggingface.co/litert-community/SmolVLM2-500M

Url of SmolVLM2-500M

Provider of SmolVLM2-500M huggingface.co

litert-community
ORGANIZATIONS

Other API from litert-community