llmrun

Best local LLMs for vision

Run local LLMs that handle images, screenshots, and docs across your GPU or Mac. We rank them using the llmrun Composite Score and highlight our top choice for every VRAM tier.

Best for vision on

8 GB12 GB16 GB24 GB32 GB48 GB+

Qwen3.8 27B

Alibaba · Q4_K_M · 17.9 GB · up to 256K context · on RTX 5090

66.9 t/s
S
How to run →

All 113 vision models for 32 GB ↓

Also worth a look

Gemma 4 31B IT

Q4_K_M · 21.7 GB · 57.1 t/s · 64K ctx

Best vision model by memory

Highest llmrun Composite Score that fits each size with at least 16K context and 10 t/s, at a Q3 or better quant. Speed is decode tokens per second on that size's reference GPU. The last row has no memory limit: the best model whatever it needs, on the GPU plan shown. Full ranking

Vision models that run on 32 GB113

Ranked by Composite Score, models without one after. Best quant that fits with a 16K context; speed is decode tokens per second on the reference GPU.

113 models · 103 run great · 10 run well

Featured only

Score: llmrun Score, 100 = average model. Fit: how well it runs on this hardware, S best to F won't fit. Ordered by Score, recency and Fit. How ranking works

Showing compatibility for NVIDIA GeForce RTX 5090

LLM models compatible with NVIDIA GeForce RTX 5090 — ranked by Score, recency and Fit
ModelVRAMScoreFit
InternVL3 5 1B
Q4_K_M·1664.0 t/s
2.2 GBS100, Runs great
SmolVLM2 500M Video Instruct
< 16K context fits
Q4_K_M·1688.1 t/s
0.9 GBS100, Runs great
MiniCPM V 4
Q4_K_M·419.0 t/s
3.1 GBS100, Runs great
Qwen3 VL 32B Instruct
Q4_K_M·56.4 t/s
23 GBA82, Runs well
UI TARS 1.5 7B
Q4_K_M·216.1 t/s
6.2 GBS100, Runs great
InternVL3 8B
Q4_K_M·222.3 t/s
7.8 GBS100, Runs great
DeepSeek OCR
< 16K context fits
Q4_K_M·728.1 t/s
2.8 GBS100, Runs great
Gemma 4 31B
Q4_K_M·57.1 t/s
21.7 GBA82, Runs well
InternVL3 1B
Q4_K_M·1878.7 t/s
2.0 GBS100, Runs great
SmolVLM 256M Instruct
< 16K context fits
Q4_K_M·2329.6 t/s
0.6 GBS100, Runs great
Qwen3 VL 4B Instruct
Q4_K_M·369.8 t/s
4.5 GBS100, Runs great
Deepseek Vl2 Tiny
BF16·205.7 t/s
9.4 GBS100, Runs great
SmolVLM2 2.2B Instruct
< 16K context fits
Q4_K_M·787.0 t/s
2.3 GBS100, Runs great
Surya Ocr 2
Q4_K_M·1532.6 t/s
0.8 GBS100, Runs great
SmolVLM 500M Instruct
< 16K context fits
Q4_K_M·1688.1 t/s
0.9 GBS100, Runs great

How much memory do you need for a vision model?

On 8 GB the best vision model is Qwen3.5 9B (score 99). At 12 GB it is Gemma 4 26B A4B IT (score 106). From 16 to 32 GB it is Qwen3.8 27B (score 122). With no memory limit, the best is DeepSeek V4.1 Flash 1017 (score 133). Add a bit more breathing room to your setup when using a vision model with an image encoder alongside the language model; each image loads as extra tokens in the context. Allocate slightly more space than you would for a text-only model of equal size.

Frequently Asked Questions

What is the best local LLM for vision?

On 24 GB, Qwen3.8 27B is the top pick with a score of 122, running at about 38 t/s on RTX 4090. It fits up to 64K tokens of context there. On 8 GB, the best fit is Qwen3.5 9B. With no memory limit, the best is DeepSeek V4.1 Flash 1017. Open a model to see which GPUs and Macs run it.

Can I run a vision model on a laptop or an 8 GB GPU?

Yes. On 8 GB the best pick is Qwen3.5 9B (Q4_K_M, 6.6 GB), which runs at about 46 t/s on RTX 5060. Laptops with 8 to 16 GB of memory can run smaller models in the same way.

Can local models read screenshots and documents?

Vision-language models take an image along with your prompt to answer questions, pull out text, or describe charts. Whether support works depends on the runtime; however, you can run them in Ollama or LM Studio.

Do images need extra memory?

The image encoder bumps up the model size. Each picture turns into hundreds of tokens within the context window, swelling the cache. Bigger images demand even more.