Instructions to use tripplet-research/taipei3.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tripplet-research/taipei3.1 with Transformers:

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

pipe = pipeline("image-text-to-text", model="tripplet-research/taipei3.1")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("tripplet-research/taipei3.1")
model = AutoModelForImageTextToText.from_pretrained("tripplet-research/taipei3.1")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use tripplet-research/taipei3.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tripplet-research/taipei3.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tripplet-research/taipei3.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/tripplet-research/taipei3.1

SGLang

How to use tripplet-research/taipei3.1 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 "tripplet-research/taipei3.1" \
    --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": "tripplet-research/taipei3.1",
		"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 "tripplet-research/taipei3.1" \
        --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": "tripplet-research/taipei3.1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use tripplet-research/taipei3.1 with Docker Model Runner:
```
docker model run hf.co/tripplet-research/taipei3.1
```

taipei3.1

File size: 1,485 Bytes

b1931c1

{
  "architectures": [
    "Mistral3ForConditionalGeneration"
  ],
  "dtype": "bfloat16",
  "image_token_index": 10,
  "model_type": "mistral3",
  "multimodal_projector_bias": false,
  "projector_hidden_act": "gelu",
  "spatial_merge_size": 2,
  "text_config": {
    "attention_dropout": 0.0,
    "bos_token_id": 1,
    "eos_token_id": 2,
    "head_dim": 128,
    "hidden_act": "silu",
    "hidden_size": 5120,
    "initializer_range": 0.02,
    "intermediate_size": 32768,
    "max_position_embeddings": 131072,
    "model_type": "mistral",
    "num_attention_heads": 32,
    "num_hidden_layers": 40,
    "num_key_value_heads": 8,
    "pad_token_id": null,
    "rms_norm_eps": 1e-05,
    "rope_parameters": {
      "rope_theta": 1000000000.0,
      "rope_type": "default"
    },
    "sliding_window": null,
    "tie_word_embeddings": false,
    "use_cache": true,
    "vocab_size": 131072
  },
  "tie_word_embeddings": true,
  "transformers_version": "5.7.0",
  "vision_config": {
    "attention_dropout": 0.0,
    "head_dim": 64,
    "hidden_act": "silu",
    "hidden_size": 1024,
    "image_size": 1540,
    "initializer_range": 0.02,
    "intermediate_size": 4096,
    "model_type": "pixtral",
    "num_attention_heads": 16,
    "num_channels": 3,
    "num_hidden_layers": 24,
    "patch_size": 14,
    "rope_parameters": {
      "rope_theta": 10000.0,
      "rope_type": "default"
    }
  },
  "vision_feature_layer": -1
}