Dots.ocr — Hardware Requirements & GPU Compatibility
Vision
dots.ocr is a 3-billion-parameter vision-language model built around a compact 1.7-billion-parameter language backbone, purpose-built for multilingual document layout parsing and OCR. It unifies layout detection and text recognition in one model, switching tasks — table extraction, formula recognition, reading order — by changing the prompt, instead of chaining separate detection models. Despite its small footprint, its developers report state-of-the-art results on document-parsing benchmarks, including strong performance on low-resource languages. It runs comfortably on a single consumer GPU. Context length is 131,072 tokens, ample for long documents. It is released under the MIT license, a highly permissive option for commercial use. Published in July 2025, it shows a compact single VLM can rival dedicated layout-detection models like DocLayout-YOLO.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 128K)Best on 24 GB (Q4_K_M · ~301 tok/s · 128K)Macs from 8 GB (Q4_K_M · ~33 tok/s)
MIT
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Dots.ocr
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/mradermacher/dots.ocr-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 21.4 GB free · 16K context
Download: dots.ocr.Q4_K_M.gguf1.12 GBmradermacher
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (9 GB headroom) · Q4_K_M
- Generation speed
- ~272 tok/s
- generation speed
- Cost per 1M output tokens
- $0.12
- per 1M output tokens
How Much VRAM Does Dots.ocr Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.6 GB | 2.5 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~397 tok/s | |
| 1.7 GB | 2.6 GB | 3-bit small — 3-bit small quantization | Fits: 128K context · ~388 tok/s | |
| 1.8 GB | 2.7 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~356 tok/s | |
| 1.9 GB | 2.8 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 128K context · ~349 tok/s | |
| 2.2 GB | 3.1 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~301 tok/s | |
| 2.5 GB | 3.4 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~260 tok/s | |
| 2.9 GB | 3.8 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~228 tok/s | |
| 3.4 GB | 4.3 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~193 tok/s |
Selected Q4_K_M
4-bit medium quantization2.6GBat 16K context
- Weights
- 1.8 GB
- KV cache
- 0.5 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 21.4 GB free · ~301 tok/s
≥50 tok/s needs ≥168 GB/s · RTX 5090 ~534, RTX 3090 Ti ~301
Which Hardware Can Run Dots.ocr?
Q4_K_M · 2.2 GBDots.ocr (Q4_K_M) requires 2.2 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 131K context window can add up to 3.7 GB, bringing total usage to 5.9 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Dots.ocr, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 5.4 GB free · up to 128K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~227Simulate output on RTX 3080RTX 2080 Ti~184Simulate output on RTX 2080 TiRTX 3070 Ti~181Simulate output on RTX 3070 TiM3*8 GB~33Simulate output on iMac 24" M3 (8 GB)Orin Nano~30Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~22Simulate output on iPhone 1720 GB
fits · 17.4 GB free · up to 128K ctx
1 GPU
RX 7900 XT~220Simulate output on RX 7900 XT32–40 GB
fits · 29.4 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~534Simulate output on RTX 5090M5 Max36 GB~148Simulate output on Mac Studio M5 Max (36 GB)M532 GB~49Simulate output on MacBook Air 13" M5 (32 GB)Orin~61Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~40Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| SmolLM3 3BHugging Face | — | 3.4 GB | ~285 | same size |
| Qwen2.5 Coder 3BAlibaba | — | 2.8 GB | ~294 | same size |
| Llama 3.2 3B InstructMeta | — | 4.1 GB | ~309 | same size |
| Starcoder2 3BBigCode | — | 2.6 GB | ~301 | same size |
| AI21 Jamba Reasoning 3BAI21 Labs | — | 2.3 GB | ~291 | same size |
| MiniMax M3 EAGLE3Inferact | — | 2.7 GB | ~284 | same size |
Specifications
- Publisher
- dots-studio
- Parameters
- 3.0B
- Architecture
- DotsOCRForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2025-07-30
- License
- MIT
Frequently Asked Questions
How much VRAM does Dots.ocr need?
Dots.ocr requires 2.2 GB of VRAM at Q4_K_M, or 6.4 GB at BF16. Full 131K context adds up to 3.7 GB (5.9 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 3.0B × 4.8 bits ÷ 8 = 1.8 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 4.1 GB (at full 131K context)
VRAM usage by quantization
What's the best quantization for Dots.ocr?
For Dots.ocr, Q4_K_M (2.2 GB) offers the best balance of quality and VRAM usage. Q5_K_S (2.5 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 1.2 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Dots.ocr on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Dots.ocr. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.2 GB of usable unified memory (RAM minus macOS overhead).
Can I run Dots.ocr locally?
Yes — Dots.ocr can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Dots.ocr?
At Q4_K_M, Dots.ocr can reach ~2202 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~301 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 2.18 × 0.65 = ~2385 tok/s
Estimated speed at Q4_K_M (2.2 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Dots.ocr?
At Q4_K_M, the download is about 1.82 GB. The full-precision BF16 version is 6.08 GB. The smallest option (IQ2_XXS) is 0.84 GB.
Which GPUs can run Dots.ocr?
52 consumer GPUs can run Dots.ocr at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Dots.ocr?
163 devices with unified memory can run Dots.ocr at Q4_K_M (2.2 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.