rednote-hilab

dots.ocr

3B parameters · Base · dots.ocr family

Long context — how it uses memory →
mit hash verified source matched revision pinned

Which version should I download?

Default pick: safetensors · BF16 — BF16 source weights (rednote-hilab/dots.ocr, MIT-based license) — reference multilingual document-OCR VLM. Community GGUFs are generic auto-quant on a custom arch; safetensors is the reliable format (Transformers/vLLM). Full repo bundled.

Set your rig to confirm it fits and to see if a higher-quality quant runs fully on your hardware.

Set your rig

Quantizations

Why these quants? Which quant should I pick? What do these mean?

safetensors · BF16

5.66 GB

Set your rig
Est. speed
—
Swarm
2S / 0L webseed OK

BF16 source weights (rednote-hilab/dots.ocr, MIT-based license) — reference multilingual document-OCR VLM. Community GGUFs are generic auto-quant on a custom arch; safetensors is the reliable format (Transformers/vLLM). Full repo bundled.

Magnet .torrent

Want a different quant? Request it on the board →

Runs fully on

RTX 5090 · BF16 RTX 5060 Ti 16GB · BF16 RTX 5070 Ti · BF16 RTX 5080 · BF16 RTX 5070 · BF16

Green = the model's best quant fits fully in GPU/unified memory at 8K context. Tap a card for its full "what runs on it" page.

Context

Advertised 128k · no independent evidence yet

No independent long-context evidence has been graded for this model yet — the advertised window above is the maintainer's number, not a usable-context claim.

Run it

Runtime completeness (BF16 torrent): llama.cpp – Ollama – vision sidecar –

  • – llama.cpp — no GGUF artifact for this torrent
  • – Ollama — not an Ollama-native artifact
  • – vision sidecar — not a vision model

Context / KV 8,192 tokens · FP16 set above the quant table

Generic commands (no rig set). GPU-offload values assume the model fits on your GPU — set your rig for values tuned to your hardware.

llama.cpp loads GGUF files only; this quant is a safetensors build.

Safetensors weights run under vLLM or transformers; a GGUF conversion, if one exists, is a separate quant row.

Ollama runs GGUF builds only; this quant is a safetensors build.

Safetensors weights run under vLLM or transformers; a GGUF conversion, if one exists, is a separate quant row.

Serve an OpenAI-compatible endpoint 
vllm serve rednote-hilab/dots.ocr --max-model-len 8192

Serves on http://localhost:8000/v1 by default.

Install 
pip install transformers accelerate torch
Load and run (run.py) 
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "rednote-hilab/dots.ocr"
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    dtype="auto",
    device_map="auto",
    trust_remote_code=True,
)

Loads the weights straight from the repo; point model_id at a local directory to use your downloaded copy. This repo ships its own modelling code, so trust_remote_code=True is required — read that code before running it.

LM Studio loads GGUF (and MLX on Apple Silicon) only; this quant is a safetensors build.

Safetensors weights run under vLLM or transformers; a GGUF conversion, if one exists, is a separate quant row.

MLX runs MLX-format weights only (Apple Silicon). This quant is a safetensors build.

Safetensors weights run under vLLM or transformers; a GGUF conversion, if one exists, is a separate quant row.

Technical details

Chat template

{%- for m in messages %}
    {%- if m.role == 'system' %}
        {{- '<|system|>' + m.content + '<|endofsystem|>\n' }}
    {%- elif m.role == 'user' %}
        {{- '<|user|>' + m.content + '<|endofuser|>' }}
    {%- elif m.role == 'assistant' %}
        {{- '<|assistant|>' + m.content }}
        {%- if not loop.last %}
            {{- '<|endofassistant|>' }}
        {%- endif %}
    {%- endif %}
{%- endfor %}
{%- if messages[-1].role != 'assistant' %}
    {{- '<|assistant|>' }}
{%- endif %}

Stop strings

<|endoftext|> <|endofassistant|>

Evidence & provenance

Source

Provenance
https://huggingface.co/rednote-hilab/dots.ocr
Revision pin
c0111ce6bc07803dbc267932ffef0ae3a51dc951
Manifest
Present

License

Name
mit
Commercial use
yes
Access
Open

How verification works →

File hashes (SHA-256)

  • dots-ocr/.gitattributes 11ad7efa24975ee4b0c3c3a38ed18737f0658a5f75a0a96787b576a78a023361
  • dots-ocr/LICENSE b34e9d9a83c0fadcbe7e3e53d850ccf9b2f99dfa2bd0cd6772fe3dfbb07c4af6
  • dots-ocr/NOTICE df6e0c47ba4b8eaf130e578a097a027ca06c8b13efd03c5539a3f6c68b0211cc
  • dots-ocr/README.md e59f00e56945936a65c74cf53333a9051715ba45e91da0b7d86cd05f5964c2eb
  • dots-ocr/chat_template.json 315c86fc560a11463ab2360aea5c7001c56533cb7229ee5cbcaaecca6d210698
  • dots-ocr/config.json 41d5c9aad72c730bb97062cab7a59cdb1d427eeec0b7a6e2f95e1ee72ad3b498
  • dots-ocr/configuration_dots.py e86a4fd70c6f21178c57112c44d5cc64b8ee02e3e8ad8cafe00045c0844b931b
  • dots-ocr/generation_config.json daaf4cfc2d4bd75bc68d07a8e51dcfb464f99e59a8d69772b25f1c793c7b3cdf
  • dots-ocr/merges.txt 599bab54075088774b1733fde865d5bd747cbcc7a547c5bc12610e874e26f5e3
  • dots-ocr/model-00001-of-00002.safetensors ea1d532184f3adf5cbcfcc00b2cf5b2abfa6fe182768a3ae63d441a9b5fc99ac
  • dots-ocr/model-00002-of-00002.safetensors 26ab1ec6c8b4e4116befbd59af42159f1dbcb0ad0c045a15e890bb2f6e8b0dae
  • dots-ocr/model.safetensors.index.json 90b4c82045a7d1343348d59c584d124a9cc0e23f3ca8b7d81250e93030924842
  • dots-ocr/modeling_dots_ocr.py 274b5bfa35f624f15001da70414e733b89f5cb44b5453d80023bff26a68d0135
  • dots-ocr/modeling_dots_ocr_vllm.py fc70aa417e8c0f5589366fbdc789193c97f4c4d0a094f867b222092188b68b19
  • dots-ocr/modeling_dots_vision.py 9fb57cce892e6f065ff86b6f02fbd72110fe6d9f749c2ae8c82897068bbed877
  • dots-ocr/preprocessor_config.json fec8b3187ab0ae340f6d0b1f6f892a53c63b73b328836c83dd7bff1d5bc4c79a
  • dots-ocr/special_tokens_map.json 92b060f39417514eed2f7652f454a3d5afb0e4a57b9f4b51dd4d1dfd8fdb9736
  • dots-ocr/tokenizer.json 386545eb05f08c51352cde2fcc2c867f1592bb330f305efd1c6a57a93b1244cd
  • dots-ocr/tokenizer_config.json 4f09a26dd6b20c82e9678d96b7363a58a778a948a21261d3315a949d47ed4cf4
  • dots-ocr/vocab.json ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910

Explore further

Appears in

What runs on 8 GB of VRAM Coding models that fit 8 GB Writing models that fit 8 GB General models that fit 8 GB Reasoning models that fit 8 GB

Also consider

Get an email when a better quant fits your rig.

Set your rig once; we'll alert you when a new or better-fitting build lands.

Set your rig

Performance reports

Real-world throughput reported by the community (and scraped sources).

View performance leaderboard
Community

Reviews, sampler presets and community runtime reports. Be the first to contribute.

Reviews (0)

Log in to write a structured review of this model.

No reviews yet.

Sampler presets

Log in to share a sampler preset or vote on presets.

No presets yet.

Community runtime reports

Unofficial, community-submitted "it loads for me" reports. These are not the official verified-working badge — an admin reviews reports before anything is marked officially verified.

Log in to report whether a quant loads in your runtime.

No community reports yet.

selected to compare · pick at least 2

Compare ()