Surya Ocr 2 — Hardware Requirements & GPU Compatibility
Vision
Surya OCR 2 is Datalab's roughly 650-million-parameter document model, built on a Qwen3.5-style vision-language architecture. One model handles OCR, layout analysis with reading order, and table recognition, producing layout JSON or full-page HTML depending on the prompt. The card reports 83.3% on olmOCR-bench, which it calls the top score under 3 billion parameters, and about 5 pages per second on an RTX 5090. At this size it runs comfortably on a modest consumer GPU. The context length is 262,144 tokens. The code is Apache 2.0, but the weights use a modified AI Pubs Open RAIL-M license that the card describes as free for research, personal use and startups under $5M in funding or revenue. Broader commercial use needs a separate license from Datalab. It was published in May 2026.
1.4M downloads
119 likesSpecifications
- Publisher
- datalab-to
- Parameters
- 686M
- Architecture
- Qwen3_5ForConditionalGeneration
- Context Length
- 262,144 tokens
- Vocabulary Size
- 65,425
- Release Date
- 2026-05-14
- License
- openrail
Get Started
HuggingFace
How Much VRAM Does Surya Ocr 2 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 0.6 GB | 7.0 GB | 0.29 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 0.7 GB | 7.1 GB | 0.33 GB | 3-bit medium quantization |
| Q4_K_Mest. | 4.80 | 0.8 GB | 7.2 GB | 0.41 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 0.8 GB | 7.2 GB | 0.49 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 0.9 GB | 7.3 GB | 0.57 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 1.0 GB | 7.4 GB | 0.69 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 1.7 GB | 8.1 GB | 1.37 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run Surya Ocr 2?
Q4_K_M · 0.8 GBSurya Ocr 2 (Q4_K_M) requires 0.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. Using the full 262K context window can add up to 6.4 GB, bringing total usage to 7.2 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run Surya Ocr 2?
Q4_K_M · 0.8 GB59 devices with unified memory can run Surya Ocr 2, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomFrequently Asked Questions
- How much VRAM does Surya Ocr 2 need?
Surya Ocr 2 requires 0.8 GB of VRAM at Q4_K_M, or 1.7 GB at BF16. Full 262K context adds up to 6.4 GB (7.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 686M × 4.8 bits ÷ 8 = 0.4 GB
KV Cache + Overhead ≈ 0.4 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 6.8 GB (at full 262K context)
VRAM usage by quantization
Q4_K_M0.8 GBQ4_K_M + full context7.2 GB- What's the best quantization for Surya Ocr 2?
For Surya Ocr 2, Q4_K_M (0.8 GB) offers the best balance of quality and VRAM usage. Q5_K_M (0.8 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 0.6 GB.
VRAM requirement by quantization
Q2_K0.6 GBQ4_K_M ★0.8 GBQ5_K_M0.8 GBQ6_K0.9 GBQ8_01.0 GBBF161.7 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Surya Ocr 2 on a Mac?
Surya Ocr 2 requires at least 0.6 GB at Q2_K, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Surya Ocr 2 locally?
Yes — Surya Ocr 2 can run locally on consumer hardware. At Q4_K_M quantization it needs 0.8 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Surya Ocr 2?
At Q4_K_M, Surya Ocr 2 can reach ~6316 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~862 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 ÷ 0.8 × 0.65 = ~6842 tok/s
Estimated speed at Q4_K_M (0.8 GB)
~6842 tok/s~862 tok/s~6842 tok/s~6316 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Surya Ocr 2?
At Q4_K_M, the download is about 0.41 GB. The full-precision BF16 version is 1.37 GB. The smallest option (Q2_K) is 0.29 GB.
- Which GPUs can run Surya Ocr 2?
52 consumer GPUs can run Surya Ocr 2 at Q4_K_M (0.8 GB). 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 Surya Ocr 2?
59 devices with unified memory can run Surya Ocr 2 at Q4_K_M (0.8 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Apple iPhone 17 Pro, Asus ROG Flow Z13 (2025, 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.