Guides / Phones and edge
Best tiny LLMs for phones and edge devices
Updated November 2026
As of November 2026, the best option is Nanbeige4.2 3B by Nanbeige (Kanzhun), with a Small LM Index of 70.4, followed by Spark-X2.6 4B (67.5) and Spark-X2.5 4B (65.5).
Models under 5B parameters, ranked by the Small LM Index. Try other hardware and context lengths in the Can I run it? tool.
| # | Model | Lab | Small LM Index | Params | Memory (4-bit) | API / 1M | License |
|---|---|---|---|---|---|---|---|
| 1 | Nanbeige4.2 3B | Nanbeige (Kanzhun) | 70.4 | 4.2B | ~3.8 GB | self-host | Apache-2.0 |
| 2 | Spark-X2.6 4B | SparkLLM (XHToken) | 67.5 | 4.1B | ~3.4 GB | self-host | Apache-2.0 |
| 3 | Spark-X2.5 4B | SparkLLM (XHToken) | 65.5 | 4.1B | ~3.4 GB | self-host | Apache-2.0 |
| 4 | LFM3 1.7B | Liquid AI | 60.0 | 1.7B | ~1.7 GB | self-host | LFM Open License |
| 5 | Nemotron Labs Dory 1.0 4B | NVIDIA | 60.0 | 4B | ~3.4 GB | self-host | OpenMDW-1.1 |
| 6 | Qwen3.5 4B | Alibaba (Qwen) | 57.8 | 4B | ~3.3 GB | self-host | Apache-2.0 |
| 7 | MiniCPM5 2B | OpenBMB (MiniCPM) | 57.5 | 2.5B | ~2.4 GB | self-host | Apache-2.0 |
| 8 | Qwen3 4B Thinking 2507 | Alibaba (Qwen) | 55.7 | 4B | ~4.2 GB | self-host | Apache-2.0 |
| 9 | Granite 4.2 3B | IBM (Granite) | 54.8 | 3B | ~3 GB | self-host | Apache-2.0 |
| 10 | Spark-X2.6 1.7B | SparkLLM (XHToken) | 50.1 | 1.7B | ~1.7 GB | self-host | Apache-2.0 |
| 11 | Ministral 3 3B | Mistral AI | 49.4 | 3B | ~3.3 GB | $0.10 | Apache-2.0 |
| 12 | LFM2.5 2.6B | Liquid AI | 43.2 | 2.7B | ~2.3 GB | $0.39 | LFM Open License |
| 13 | EXAONE 4.0 1.2B | LG AI Research (EXAONE) | 36.1 | 1.3B | ~1.8 GB | self-host | EXAONE License (non-commercial) |
| 14 | Qwen3.5 2B | Alibaba (Qwen) | 33.0 | 2.3B | ~2 GB | self-host | Apache-2.0 |
| 15 | SmolLM3 3B | Hugging Face (SmolLM) | 30.4 | 3.1B | ~3 GB | self-host | Apache-2.0 |
Memory covers the weights only; long contexts need more. Scores are mostly lab-reported, so gaps of 1–2 points are noise. See the methodology, or filter the full leaderboard yourself.
Head to head
Questions
What is the best small language model for phones and edge devices?
As of November 2026, the best option is Nanbeige4.2 3B by Nanbeige (Kanzhun), with a Small LM Index of 70.4, followed by Spark-X2.6 4B (67.5) and Spark-X2.5 4B (65.5).
How is this list ranked?
Models under 5B parameters, ranked by the Small LM Index.
Which of the top five needs the least memory?
LFM3 1.7B, at about 1.7 GB with 4-bit quantization.
Which is the cheapest to use through an API?
Ministral 3 3B, at $0.10 per million tokens from the cheapest provider on OpenRouter.
Other guides
- Best small open models for coding
- Best small open models for math
- Best LLMs that run on an 8 GB GPU or laptop
- Best LLMs that run on a 16 GB GPU
- Best LLMs for a 24 GB GPU (RTX 4090, RTX 3090)
- Best cheap LLM APIs (open models)
- Best small open vision-language models
- Best open models with an Apache 2.0 or MIT license
- Best European open language models
- Best American open models under 100B
- Best fully open models (open data and code)