SmolLM-135M
SmolLM is a series of state-of-the-art small language models available in three sizes: 135M, 360M, and 1.7B parameters. These models are built on Cosmo-Corpus, a meticulously curated high-quality training dataset. Cosmo-Corpus includes Cosmopedia v2 (28B tokens of synthetic textbooks and stories generated by Mixtral), Python-Edu (4B tokens of educational Python samples from The Stack), and FineWeb-Edu (220B tokens of...
Params
130 M
Context
2,048
Downloads 30d
293 K
Likes
263
Download history
daily snapshots · 10 days293 K243 K
Jul 28Jul 31Aug 3Aug 6
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | f32 | 0.5 GB | 1.1 GB | ✅ Runs comfortably |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ curl -s https://aimodelscomparison.com/api/v1/models/smollm-135m
{
"hf_id": "HuggingFaceTB/SmolLM-135M",
"params_b": 0.13,
"context_length": 2048,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 292615,
"vram_estimates": [
{ "quant": "f32", "gb": 1.1 }
],
"updated_at": "2026-07-28T18:07:22Z"
}
Specifications
- Architecture
- LlamaForCausalLM
- Parameters
- 130 M
- Tensor type
- F32
- Context length
- 2,048
- Vocabulary
- 49,152
- Layers / heads
- 30 / 9
- Licence
- apache-2.0
- First seen on the Hub
- 2024-07-14
- Training datasets
- HuggingFaceTB/smollm-corpus
- Added to our catalog
- 2026-07-28
Compare with
Sponsored · GPU cloud
Not enough VRAM?
Spin up a 24 GB L4 instance in 40 seconds. $0.44/hr.