TeleOCR — Hardware Requirements & GPU Compatibility
Vision
TeleOCR is a lightweight vision-language model of about 1.2 billion parameters, released under the XingChen-AGI organization and designed for document parsing in Chinese and English. Its distinguishing point, according to the card, is that it handles both digital documents and camera-captured ones within a single framework, using geometry-aware document modeling and a progressive four-stage training pipeline. The card says it was renamed from NaviDC-OCR in September 2026. At this size, it fits on a single modest consumer GPU or even a CPU once quantized. The context window is 128,000 tokens. It is released under the Apache 2.0 license, permitting unrestricted commercial and research use. Published in August 2026, the card reports it outperforming MinerU 2.5 Pro and PaddleOCR-VL 1.6 on the Dr.DocBench challenge evaluation, and a community GGUF conversion exists for llama.cpp.
32.7K downloads
1.3K likesSpecifications
- Publisher
- XingChen-AGI
- Parameters
- 1.4B
- Architecture
- Qwen2_5_VLForConditionalGeneration
- Context Length
- 128,000 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2026-08-14
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does TeleOCR Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 1.0 GB | 8.2 GB | 0.60 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 1.1 GB | 8.3 GB | 0.69 GB | 3-bit medium quantization |
| Q4_K_Mest. | 4.80 | 1.3 GB | 8.5 GB | 0.85 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 1.4 GB | 8.7 GB | 1.01 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 1.6 GB | 8.8 GB | 1.17 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 1.8 GB | 9.1 GB | 1.42 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 3.3 GB | 10.5 GB | 2.83 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 TeleOCR?
Q4_K_M · 1.3 GBTeleOCR (Q4_K_M) requires 1.3 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 128K context window can add up to 7.2 GB, bringing total usage to 8.5 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run TeleOCR?
Q4_K_M · 1.3 GB59 devices with unified memory can run TeleOCR, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomWhere to Download TeleOCR
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 TeleOCR need?
TeleOCR requires 1.3 GB of VRAM at Q4_K_M, or 3.3 GB at BF16. Full 128K context adds up to 7.2 GB (8.5 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 1.4B × 4.8 bits ÷ 8 = 0.8 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 7.7 GB (at full 128K context)
VRAM usage by quantization
Q4_K_M1.3 GBQ4_K_M + full context8.5 GB- What's the best quantization for TeleOCR?
For TeleOCR, Q4_K_M (1.3 GB) offers the best balance of quality and VRAM usage. Q5_K_M (1.4 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.0 GB.
VRAM requirement by quantization
Q2_K1.0 GBQ4_K_M ★1.3 GBQ5_K_M1.4 GBQ6_K1.6 GBQ8_01.8 GBBF163.3 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run TeleOCR on a Mac?
TeleOCR requires at least 1.0 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 TeleOCR locally?
Yes — TeleOCR can run locally on consumer hardware. At Q4_K_M quantization it needs 1.3 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is TeleOCR?
At Q4_K_M, TeleOCR can reach ~3780 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~516 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 ÷ 1.3 × 0.65 = ~4095 tok/s
Estimated speed at Q4_K_M (1.3 GB)
~4095 tok/s~516 tok/s~4095 tok/s~3780 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of TeleOCR?
At Q4_K_M, the download is about 0.85 GB. The full-precision BF16 version is 2.83 GB. The smallest option (Q2_K) is 0.60 GB.
- Which GPUs can run TeleOCR?
52 consumer GPUs can run TeleOCR at Q4_K_M (1.3 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 TeleOCR?
59 devices with unified memory can run TeleOCR at Q4_K_M (1.3 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.