Image-Text-to-Text
ZeroModels
Keras
PyTorch
JAX
TensorFlow
English
internvl
internvl3-5
multimodal
vision
Instructions to use zeromodels/internvl3.5-2b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ZeroModels
How to use zeromodels/internvl3.5-2b with ZeroModels:
# pip install -U zeromodels # ZeroModels is pure Keras 3, so pick a backend: "jax", "torch" or "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" from zeromodels import AutoZModel # AutoZModel reads the repo's model_type and loads the matching class. # For a task head use the matching loader, e.g. AutoZMImageClassify / AutoZMDetect / # AutoZMSemanticSegment / AutoZMTextGenerate (see zeromodels.auto). model = AutoZModel.from_weights("zeromodels/internvl3.5-2b") - Keras
How to use zeromodels/internvl3.5-2b with Keras:
# !pip install -U keras tensorflow huggingface_hub # Keras needs TensorFlow installed to read "hf://" paths, so the tensorflow backend is selected here; # "jax" and "torch" also work for computation once TensorFlow is installed. import os os.environ["KERAS_BACKEND"] = "tensorflow" import keras model = keras.saving.load_model("hf://zeromodels/internvl3.5-2b") - Notebooks
- Google Colab
- Kaggle
Download zm_config.json from zeromodels/internvl3.5-2b: direct link, hf CLI and curl.
- Browser
- Download file 1.1 kB
-
https://huggingface.co/zeromodels/internvl3.5-2b/resolve/main/zm_config.json
- Command line
-
hf download hf://zeromodels/internvl3.5-2b/zm_config.json
-
curl -L -o zm_config.json https://huggingface.co/zeromodels/internvl3.5-2b/resolve/main/zm_config.json
1.1 kB
| { | |
| "library_name": "zeromodels", | |
| "zeromodels_version": "1.2.4", | |
| "model_module": "zeromodels.models.internvl", | |
| "model_class": "InternVLConditionalGenerate", | |
| "variant": "internvl3.5-2b", | |
| "weights": "model.weights.h5", | |
| "schema_version": 2, | |
| "weight_dtype": "bfloat16", | |
| "model_type": "internvl", | |
| "text_config": { | |
| "vocab_size": 151936, | |
| "embed_dim": 2048, | |
| "mlp_dim": 6144, | |
| "num_layers": 28, | |
| "num_heads": 16, | |
| "num_kv_heads": 8, | |
| "head_dim": 128, | |
| "text_backbone": "qwen3", | |
| "norm_eps": 1e-06, | |
| "rope_theta": 1000000, | |
| "tie_embeddings": false, | |
| "num_experts": 0, | |
| "num_experts_per_tok": 0, | |
| "moe_mlp_dim": 0, | |
| "norm_topk_prob": true, | |
| "decoder_sparse_step": 1, | |
| "mlp_only_layers": [] | |
| }, | |
| "vision_config": { | |
| "embed_dim": 1024, | |
| "mlp_dim": 4096, | |
| "num_layers": 24, | |
| "num_heads": 16, | |
| "image_size": 448, | |
| "patch_size": 14, | |
| "attention_bias": true, | |
| "qk_norm": false, | |
| "norm_type": "layer_norm", | |
| "norm_eps": 1e-06, | |
| "layer_scale_init": 0.1 | |
| }, | |
| "downsample_ratio": 0.5, | |
| "image_token_id": 151671 | |
| } | |