Instructions to use lattice-nlp/coref-llm-CRAC26-multilingual with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lattice-nlp/coref-llm-CRAC26-multilingual with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("google/gemma-3-27b-it")
model = PeftModel.from_pretrained(base_model, "lattice-nlp/coref-llm-CRAC26-multilingual")

Transformers

How to use lattice-nlp/coref-llm-CRAC26-multilingual with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="lattice-nlp/coref-llm-CRAC26-multilingual")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("lattice-nlp/coref-llm-CRAC26-multilingual", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lattice-nlp/coref-llm-CRAC26-multilingual with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lattice-nlp/coref-llm-CRAC26-multilingual"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lattice-nlp/coref-llm-CRAC26-multilingual",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lattice-nlp/coref-llm-CRAC26-multilingual

SGLang

How to use lattice-nlp/coref-llm-CRAC26-multilingual with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "lattice-nlp/coref-llm-CRAC26-multilingual" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lattice-nlp/coref-llm-CRAC26-multilingual",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "lattice-nlp/coref-llm-CRAC26-multilingual" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lattice-nlp/coref-llm-CRAC26-multilingual",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lattice-nlp/coref-llm-CRAC26-multilingual with Docker Model Runner:
```
docker model run hf.co/lattice-nlp/coref-llm-CRAC26-multilingual
```

Model Card for coref-llm-CRAC26-multilingual

This model is a fine-tuned version of google/gemma-3-27b-it. It has been trained using TRL.

CRAC 2026 Shared Task — Test Set Results

Dataset	F1 Score
cs_pcedt	72.79
hu_korkor	60.54
en_gum	73.88
la_coreflat	33.56
cs_pdt	76.31
no_bokmaalnarc	81.19
cs_pdtsc	70.89
fr_democrat	56.76
ca_ancora	78.00
pl_pcc	78.35
cu_proiel	61.32
nl_openboek	77.80
hi_hdtb	73.74
hu_szegedkoref	62.81
ko_ecmt	68.10
hbo_ptnk	75.65
lt_lcc	60.45
tr_itcc	68.39
grc_proiel	74.11
fr_ancor	77.33
en_litbank	81.56
no_nynorsknarc	79.35
es_ancora	78.30
fr_litbankfr	73.34
de_potsdamcc	70.87
en_fantasycoref	78.47
ru_rucor	81.99
Average	71.33