chandra-ocr-2
datalab-toopenrailtextvision

chandra-ocr-2

Updated Nov 21, 2026
Parameters
5.3B
Context
262,144
License
openrail
Q4 GGUF
3.07 GB
Updated
Nov 21, 2026

Q4 GGUF download: mradermacher/chandra-ocr-2-GGUF - 3.07 GB on disk, read 2026-11-14. VRAM figures above are estimates; this is the actual file size.

Will it run on your hardware?

Pick your GPU memory - see which quantizations fit, and the cheapest card for the rest

Runs on your 24 GB - best at FP16
4 of 4 quantizations fit chandra-ocr-2 with real runtime overhead.
Q4_K_M
3 GB
Q5_K_M
4 GB
Q8_0
6 GB
FP16
11 GB
fits tight too big

Need an exact figure for your context length? Use the VRAM calculator.

Run it locally

Copy-paste - running in under a minute

vLLMOpenAI-compatible API
vllm serve datalab-to/chandra-ocr-2

New to this? Start with Ollama · serve to many users with vLLM.

Deep dive

Notes, sources, and the full write-up

chandra-ocr-2 is an open-weight vision-language model published by datalab-to on Hugging Face in March 2026, with 5.3B parameters under the openrail license.

Among the 72 vision-language models in this directory it ranks 18th by monthly Hugging Face downloads, and 55 of the others are larger.

HardwareLink to this section

Unquantized FP16 weights take about 11 GB; 8-bit brings that to ~6 GB and 4-bit (Q4_K_M) to ~3 GB, before activations and any context or image buffers. A 12 GB consumer card holds it at 4-bit; FP16 needs a 24 GB card such as an RTX 4090. Check exact numbers for your card with the VRAM calculator.

Similar vision-language modelsLink to this section

PopularityLink to this section

chandra-ocr-2 was downloaded 1,339,389 times on Hugging Face in the last month (14,559,550 all-time) and has 610 likes.

Run it locallyLink to this section

Weights and usage instructions are on the datalab-to/chandra-ocr-2 model card; it loads with Hugging Face Transformers.

Local runtimes that fit this model:

  • vllm - high-throughput serving on NVIDIA GPUs
  • transformers - the reference Python library most model cards target

Frequently asked

Quick answers to common questions

How much VRAM does chandra-ocr-2 need?

chandra-ocr-2 with 5.3B parameters needs approximately 3 GB at Q4_K_M quantization. Use our VRAM calculator for an exact estimate.

Is chandra-ocr-2 better than other datalab-to models?

chandra-ocr-2 has 5.3B parameters with 262,144 context - a strong choice for vision, long-context.

What license is chandra-ocr-2 under?

chandra-ocr-2 is released under the openrail license, making it suitable for most commercial and personal projects.

What hardware runs chandra-ocr-2 well?

With 5.3B parameters, chandra-ocr-2 requires adequate VRAM. High-end GPUs like the RTX 4090 (24GB), RTX 5090 (32GB), or Mac Studio with unified memory are good options. Check our hardware directory for specific recommendations.

What is the best quantization for chandra-ocr-2?

Q4_K_M is the recommended sweet spot - ~98% of FP16 quality at ~27% of the size. Q5_K_M (~4 GB) is an option if you have spare VRAM. Use our VRAM calculator to compare.

How long can chandra-ocr-2's context window handle?

chandra-ocr-2 supports a 262,144-token context window - enough for very long documents, codebases, or multi-turn conversations. Real-world usable context may vary by implementation.

What models compete with chandra-ocr-2?

chandra-ocr-2 competes with other models in its class. Browse our model directory for comparisons, benchmarks, and community reviews to find the best fit.

Nearby options

Similar models and compatible hardware by spec