MinerU2.5 Pro 2604 1.2B — Hardware Requirements & GPU Compatibility
Vision
MinerU2.5-Pro-2604-1.2B is opendatalab's 1.2-billion-parameter document parsing model, built on a Qwen2-VL vision-language architecture and specialized for PDF-to-Markdown conversion rather than general chat. It handles complex layouts, tables, and formulas, and the card reports it beating both specialized OCR models like GLM-OCR and PaddleOCR-VL-1.5 and much larger general vision-language models such as Gemini 3 Pro and Qwen3-VL-235B on the OmniDocBench v1.6 leaderboard. The gains over the earlier MinerU 2.5 baseline come entirely from improved training data rather than a change to the 1.2B-parameter architecture. At this size, it fits comfortably on a single modest consumer GPU or even a CPU. Context length is 8,192 tokens. It is released under the Apache 2.0 license, permitting unrestricted commercial and research use, and was published in April 2026.
196.5K downloads
169 likesSpecifications
- Publisher
- opendatalab
- Parameters
- 1.2B
- Architecture
- Qwen2VLForConditionalGeneration
- Context Length
- 8,192 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2026-04-02
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does MinerU2.5 Pro 2604 1.2B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 0.8 GB | 0.9 GB | 0.49 GB | 2-bit quantization with K-quant improvements |
| Q3_K_S | 3.50 | 0.8 GB | 0.9 GB | 0.51 GB | 3-bit small quantization |
| Q3_K_M | 3.90 | 0.9 GB | 1.0 GB | 0.56 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 0.9 GB | 1.0 GB | 0.58 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 1.0 GB | 1.1 GB | 0.69 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 1.1 GB | 1.2 GB | 0.82 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 1.3 GB | 1.4 GB | 0.95 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 1.5 GB | 1.6 GB | 1.16 GB | 8-bit quantization, near-lossless |
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 MinerU2.5 Pro 2604 1.2B?
Q4_K_M · 1.0 GBMinerU2.5 Pro 2604 1.2B (Q4_K_M) requires 1.0 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 2+ GB is recommended. Using the full 8K context window can add up to 0.1 GB, bringing total usage to 1.1 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run MinerU2.5 Pro 2604 1.2B?
Q4_K_M · 1.0 GB59 devices with unified memory can run MinerU2.5 Pro 2604 1.2B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomWhere to Download MinerU2.5 Pro 2604 1.2B
Community quantizations of this model — GGUF for llama.cpp, Ollama, and LM Studio, plus AWQ/MLX variants where available.
Related Models
Frequently Asked Questions
- How much VRAM does MinerU2.5 Pro 2604 1.2B need?
MinerU2.5 Pro 2604 1.2B requires 1.0 GB of VRAM at Q4_K_M, or 2.6 GB at BF16.
VRAM = Weights + KV Cache + Overhead
Weights = 1.2B × 4.8 bits ÷ 8 = 0.7 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 0.4 GB (at full 8K context)
VRAM usage by quantization
Q4_K_M1.0 GBQ4_K_M + full context1.1 GB- What's the best quantization for MinerU2.5 Pro 2604 1.2B?
For MinerU2.5 Pro 2604 1.2B, Q4_K_M (1.0 GB) offers the best balance of quality and VRAM usage. Q5_K_S (1.1 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 0.6 GB.
VRAM requirement by quantization
IQ2_XXS0.6 GBIQ3_XS0.8 GBQ4_00.9 GBIQ4_NL1.0 GBQ4_K_M ★1.0 GBBF162.6 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run MinerU2.5 Pro 2604 1.2B on a Mac?
MinerU2.5 Pro 2604 1.2B requires at least 0.6 GB at IQ2_XXS, 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 MinerU2.5 Pro 2604 1.2B locally?
Yes — MinerU2.5 Pro 2604 1.2B can run locally on consumer hardware. At Q4_K_M quantization it needs 1.0 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is MinerU2.5 Pro 2604 1.2B?
At Q4_K_M, MinerU2.5 Pro 2604 1.2B can reach ~4706 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~642 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 ÷ 1.0 × 0.65 = ~5098 tok/s
Estimated speed at Q4_K_M (1.0 GB)
~5098 tok/s~642 tok/s~5098 tok/s~4706 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of MinerU2.5 Pro 2604 1.2B?
At Q4_K_M, the download is about 0.69 GB. The full-precision BF16 version is 2.31 GB. The smallest option (IQ2_XXS) is 0.32 GB.
- Which GPUs can run MinerU2.5 Pro 2604 1.2B?
52 consumer GPUs can run MinerU2.5 Pro 2604 1.2B at Q4_K_M (1.0 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 MinerU2.5 Pro 2604 1.2B?
59 devices with unified memory can run MinerU2.5 Pro 2604 1.2B at Q4_K_M (1.0 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.