Guides / Vision (image input)
Best small open vision-language models
Updated November 2026
As of November 2026, the best option is Qwen3.8 27B by Alibaba (Qwen), with a Small LM Index of 77.0, followed by Ternary Bonsai 2 27B (74.6) and Qwen3.6 27B (74.0).
Models that accept images as input, ranked by the Small LM Index. MMMU-Pro, a multimodal benchmark, is shown where reported.
| # | Model | Lab | Small LM Index | Params | Memory (4-bit) | API / 1M | License |
|---|---|---|---|---|---|---|---|
| 1 | Qwen3.8 27B | Alibaba (Qwen) | 77.0 | 27B | ~17.9 GB | $0.13 | Apache-2.0 |
| 2 | Ternary Bonsai 2 27B | PrismML | 74.6 | 27.4B | ~18.2 GB | self-host | Apache-2.0 |
| 3 | Qwen3.6 27B | Alibaba (Qwen) | 74.0 | 27B | ~17.9 GB | $0.90 | Apache-2.0 |
| 4 | Nemotron 3.5 Bolt 74B-A9B | NVIDIA | 73.3 | 74B (9B active) | ~46.9 GB | self-host | OpenMDW-1.1 |
| 5 | Muse Glimmer 30B | Meta | 72.6 | 29.6B | ~19.2 GB | $0.53 | Apache-2.0 |
| 6 | Qwen3.6 35B-A3B | Alibaba (Qwen) | 72.2 | 35B (3B active) | ~22.6 GB | $0.031 | Apache-2.0 |
| 7 | Qwen3.5 27B | Alibaba (Qwen) | 70.6 | 27B | ~17.9 GB | $0.69 | Apache-2.0 |
| 8 | Gemma 4 31B | 70.1 | 30.7B | ~21.9 GB | $0.16 | Apache-2.0 | |
| 9 | Qwen3.5 35B-A3B | Alibaba (Qwen) | 69.9 | 35B (3B active) | ~22.6 GB | $0.24 | Apache-2.0 |
| 10 | Gemma 4 26B-A4B | 67.5 | 25.2B (3.8B active) | ~16.8 GB | $0.068 | Apache-2.0 | |
| 11 | Qwen3.5 9B | Alibaba (Qwen) | 63.9 | 9B | ~6.4 GB | $0.028 | Apache-2.0 |
| 12 | Edge0 35B-A3B (preview) | Edge0 | 62.8 | 34.7B (3B active) | ~22.4 GB | self-host | Apache-2.0 |
| 13 | Apriel 1.6 15B Thinker | ServiceNow (Apriel) | 60.9 | 14.9B | ~11.4 GB | self-host | MIT |
| 14 | Ministral 3 14B | Mistral AI | 60.0 | 14B | ~10.6 GB | $0.20 | Apache-2.0 |
| 15 | Qwen3.5 4B | Alibaba (Qwen) | 57.8 | 4B | ~3.3 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 open model that understands images?
As of November 2026, the best option is Qwen3.8 27B by Alibaba (Qwen), with a Small LM Index of 77.0, followed by Ternary Bonsai 2 27B (74.6) and Qwen3.6 27B (74.0).
How is this list ranked?
Models that accept images as input, ranked by the Small LM Index. MMMU-Pro, a multimodal benchmark, is shown where reported.
Which of the top five needs the least memory?
Qwen3.8 27B, at about 17.9 GB with 4-bit quantization.
Which is the cheapest to use through an API?
Qwen3.5 9B, at $0.028 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 tiny LLMs for phones and edge devices
- Best cheap LLM APIs (open 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)