Can the RTX 3090 run Unlimited-OCR?
Yes. Unlimited-OCR fits the RTX 3090's 24 GB at Q4_K_M (about 3.8 GB needed), running at an estimated 274 tok/s. There's room for up to FP16 if you want higher quality.
Every quantization
Unlimited-OCR (3.3B) on 24 GB
| Quant | Weights | Needs | Fit | Est. speed |
|---|---|---|---|---|
| Q4_K_M | 2 GB | 3.8 GB | Fits easily | ~274 tok/s |
| Q5_K_M | 2 GB | 3.8 GB | Fits easily | ~274 tok/s |
| Q8_0 | 4 GB | 5.8 GB | Fits easily | ~137 tok/s |
| FP16 | 7 GB | 8.8 GB | Fits easily | ~78 tok/s |
Estimates. Needs = weights + 1.8 GB for KV cache and runtime at modest context. Speed is single-stream decode scaled from the RTX 3090's 936 GB/s bandwidth (XiongjieDai/GPU-Benchmarks-on-LLM-Inference). Long contexts need more memory and run slower.
Compare
Same model, other hardware - and other models on this card
Unlimited-OCR on other hardware
- RTX 5090~514 tok/s
- RTX 5090 D~513 tok/s
- RTX 6000D~450 tok/s
- RTX 4090~313 tok/s
- Radeon RX 7900 XTX~275 tok/s
- RTX 5080~275 tok/s
- RTX 3080 Ti~261 tok/s
- RTX 5080 Mobile~257 tok/s
Also runs on the RTX 3090
- Qwen3-Coder-30B-A3B-Instruct~206 tok/s
- Qwen3-30B-A3B-Instruct-2507~206 tok/s
- Qwen3.6 27B~34 tok/s
- Qwen3.8 27B~34 tok/s
- Gemma 4 26B A4B~161 tok/s
- gpt-oss-20b~182 tok/s
- Nemotron 3.5 Lightning~55 tok/s
- Qwen2.5-Coder-14B-Instruct~61 tok/s
Frequently asked
Quick answers
Can the NVIDIA RTX 3090 run Unlimited-OCR?
Yes. Unlimited-OCR fits the RTX 3090's 24 GB at Q4_K_M (about 3.8 GB needed), running at an estimated 274 tok/s. There's room for up to FP16 if you want higher quality.
How much VRAM does Unlimited-OCR need?
About 2 GB of weights at Q4_K_M, 4 GB at Q8_0 and 7 GB at FP16. Add roughly 1.8 GB for the KV cache and runtime at modest context lengths.
How fast is Unlimited-OCR on the RTX 3090?
Roughly 274 tokens per second at Q4_K_M (single user, llama.cpp-style decode). This is an estimate from the RTX 3090's 936 GB/s memory bandwidth.