Image-Text-to-Text
Transformers
Safetensors
grt_vl
text-generation
grt
grt-vl
vision-language
multimodal
rotational-transformer
givens-rotations
clifford-algebra
matterwave
Instructions to use MatterWaveAI/grt-30m-vl with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MatterWaveAI/grt-30m-vl with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="MatterWaveAI/grt-30m-vl")# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForSeq2SeqLM model = AutoModelForSeq2SeqLM.from_pretrained("MatterWaveAI/grt-30m-vl", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use MatterWaveAI/grt-30m-vl with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MatterWaveAI/grt-30m-vl" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MatterWaveAI/grt-30m-vl", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/MatterWaveAI/grt-30m-vl
- SGLang
How to use MatterWaveAI/grt-30m-vl 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 "MatterWaveAI/grt-30m-vl" \ --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": "MatterWaveAI/grt-30m-vl", "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 "MatterWaveAI/grt-30m-vl" \ --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": "MatterWaveAI/grt-30m-vl", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use MatterWaveAI/grt-30m-vl with Docker Model Runner:
docker model run hf.co/MatterWaveAI/grt-30m-vl
Download config.json from MatterWaveAI/grt-30m-vl: direct link, hf CLI and curl.
- Browser
- Download file 2.93 kB
-
https://huggingface.co/MatterWaveAI/grt-30m-vl/resolve/main/config.json
- Command line
-
hf download hf://MatterWaveAI/grt-30m-vl/config.json
-
curl -L -o config.json https://huggingface.co/MatterWaveAI/grt-30m-vl/resolve/main/config.json
2.93 kB
| { | |
| "architectures": [ | |
| "GRTVLForConditionalGeneration" | |
| ], | |
| "dtype": "float32", | |
| "image_token_id": 151655, | |
| "model_type": "grt_vl", | |
| "text_config": { | |
| "_name_or_path": "C:\\Users\\David\\Notes\\matterwaveai\\release\\30m", | |
| "angle_init_scale": 0.1, | |
| "architectures": [ | |
| "GRTForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "chunk_size_feed_forward": 0, | |
| "coupling_init_mode": "diagonal", | |
| "coupling_init_scale": 0.01, | |
| "dtype": "float32", | |
| "eos_token_id": 151643, | |
| "ffn_n_rotations": 21, | |
| "hidden_size": 256, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 1024, | |
| "is_encoder_decoder": false, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "learnable_residual_gates": true, | |
| "learnable_rope": true, | |
| "log_magnitude_init": 0.0, | |
| "max_position_embeddings": 2048, | |
| "model_type": "grt", | |
| "n_mixer_rotations": 1, | |
| "n_qkv_rotations": 11, | |
| "num_attention_heads": 4, | |
| "num_hidden_layers": 12, | |
| "output_attentions": false, | |
| "output_hidden_states": false, | |
| "output_scale_init": 30.0, | |
| "pad_token_id": 151643, | |
| "peri_ln": false, | |
| "phase_init_scale": 0.1, | |
| "pos_freq_base": 10000.0, | |
| "problem_type": null, | |
| "residual_gate_mode": "bounded_depth", | |
| "residual_norm_mode": "none", | |
| "return_dict": true, | |
| "tie_word_embeddings": true, | |
| "use_cache": true, | |
| "vocab_size": 151669 | |
| }, | |
| "tie_word_embeddings": true, | |
| "transformers_version": "5.21.0.dev0", | |
| "vision_config": { | |
| "_name_or_path": "", | |
| "angle_init_scale": 0.1, | |
| "architectures": null, | |
| "attention_dropout": 0.0, | |
| "chunk_size_feed_forward": 0, | |
| "coupling_init_mode": "diagonal", | |
| "coupling_init_scale": 0.01, | |
| "dtype": null, | |
| "ffn_n_rotations": 21, | |
| "hidden_size": 256, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "image_size": 224, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 1024, | |
| "is_causal": false, | |
| "is_encoder_decoder": false, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "learnable_residual_gates": true, | |
| "learnable_rope": true, | |
| "merger_n_rotations": 21, | |
| "model_type": "grt_vision", | |
| "n_mixer_rotations": 1, | |
| "n_qkv_rotations": 11, | |
| "num_attention_heads": 4, | |
| "num_channels": 3, | |
| "num_hidden_layers": 6, | |
| "output_attentions": false, | |
| "output_hidden_states": false, | |
| "patch_n_rotations": 11, | |
| "patch_size": 14, | |
| "peri_ln": false, | |
| "pos_freq_base": 10000.0, | |
| "problem_type": null, | |
| "residual_gate_mode": "bounded_depth", | |
| "residual_norm_mode": "none", | |
| "return_dict": true, | |
| "spatial_merge_size": 2 | |
| }, | |
| "vision_end_token_id": 151653, | |
| "vision_start_token_id": 151652 | |
| } | |