Instructions to use hf-internal-testing/pixtral-12b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hf-internal-testing/pixtral-12b with Transformers:

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

pipe = pipeline("image-text-to-text", model="hf-internal-testing/pixtral-12b")

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("hf-internal-testing/pixtral-12b")
model = AutoModelForImageTextToText.from_pretrained("hf-internal-testing/pixtral-12b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use hf-internal-testing/pixtral-12b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hf-internal-testing/pixtral-12b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hf-internal-testing/pixtral-12b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/hf-internal-testing/pixtral-12b

SGLang

How to use hf-internal-testing/pixtral-12b 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 "hf-internal-testing/pixtral-12b" \
    --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": "hf-internal-testing/pixtral-12b",
		"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 "hf-internal-testing/pixtral-12b" \
        --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": "hf-internal-testing/pixtral-12b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use hf-internal-testing/pixtral-12b with Docker Model Runner:
```
docker model run hf.co/hf-internal-testing/pixtral-12b
```

config_update

by varbspkrl - opened Sep 18, 2024

base: refs/heads/main

←

from: refs/pr/2

Discussion Files changed

-0

varbspkrl

Sep 18, 2024

•

edited Sep 18, 2024

weights shape mismatch

Script:

from transformers import LlavaForConditionalGeneration, AutoProcessor
from PIL import Image

model_id = "hf-internal-testing/pixtral-12b"
model = LlavaForConditionalGeneration.from_pretrained(model_id)
processor = AutoProcessor.from_pretrained(model_id)

IMG_URLS = [
    "https://picsum.photos/id/237/400/300",
    "https://picsum.photos/id/231/200/300",
    "https://picsum.photos/id/27/500/500",
    "https://picsum.photos/id/17/150/600",
]
PROMPT = "<s>[INST]Describe the images.\n[IMG][IMG][IMG][IMG][/INST]"

inputs = processor(images=IMG_URLS, text=PROMPT, return_tensors="pt")
generate_ids = model.generate(**inputs, max_new_tokens=500)
ouptut = processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]

Error:

Loading checkpoint shards: 100%|████████████████████████████████████████████████████████████████████| 6/6 [00:30<00:00,  5.16s/it]
Traceback (most recent call last):
  File "/Users/varb/exo/test.py", line 70, in <module>
    model = LlavaForConditionalGeneration.from_pretrained(model_id)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/varb/transformers/src/transformers/modeling_utils.py", line 3976, in from_pretrained
    ) = cls._load_pretrained_model(
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/varb/transformers/src/transformers/modeling_utils.py", line 4511, in _load_pretrained_model
    raise RuntimeError(f"Error(s) in loading state_dict for {model.__class__.__name__}:\n\t{error_msg}")
RuntimeError: Error(s) in loading state_dict for LlavaForConditionalGeneration:
        size mismatch for language_model.model.layers.0.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.0.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.0.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.0.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.1.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.1.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.1.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.1.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.2.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.2.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.2.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.2.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.3.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.3.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.3.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.3.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.4.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.4.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.4.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.4.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.5.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.5.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.5.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.5.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.6.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.6.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.6.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.6.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.7.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.7.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.7.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.7.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.8.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.8.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.8.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.8.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.9.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.9.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.9.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.9.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.10.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.10.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.10.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.10.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.11.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.11.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.11.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.11.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.12.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.12.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.12.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.12.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.13.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.13.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.13.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.13.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.14.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.14.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.14.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.14.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.15.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.15.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.15.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.15.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.16.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.16.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.16.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.16.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.17.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.17.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.17.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.17.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.18.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.18.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.18.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.18.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.19.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.19.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.19.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.19.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.20.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.20.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.20.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.20.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.21.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.21.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.21.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.21.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.22.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.22.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.22.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.22.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.23.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.23.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.23.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.23.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.24.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.24.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.24.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.24.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.25.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.25.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.25.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.25.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.26.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.26.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.26.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.26.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.27.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.27.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.27.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.27.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.28.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.28.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.28.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.28.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.29.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.29.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.29.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.29.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.30.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.30.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.30.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.30.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.31.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.31.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.31.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.31.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.32.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.32.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.32.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.32.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.33.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.33.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.33.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.33.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.34.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.34.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.34.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.34.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.35.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.35.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.35.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.35.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.36.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.36.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.36.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.36.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.37.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.37.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.37.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.37.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.38.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.38.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.38.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.38.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.39.self_attn.q_proj.weight: copying a param with shape torch.Size([4096, 5120]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        size mismatch for language_model.model.layers.39.self_attn.k_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.39.self_attn.v_proj.weight: copying a param with shape torch.Size([1024, 5120]) from checkpoint, the shape in current model is torch.Size([1280, 5120]).
        size mismatch for language_model.model.layers.39.self_attn.o_proj.weight: copying a param with shape torch.Size([5120, 4096]) from checkpoint, the shape in current model is torch.Size([5120, 5120]).
        You may consider adding `ignore_mismatched_sizes=True` in the model `from_pretrained` method.

Updated Script:

from transformers import LlavaForConditionalGeneration, AutoProcessor
from PIL import Image

model_id = "hf-internal-testing/pixtral-12b"
model = LlavaForConditionalGeneration.from_pretrained(model_id)
processor = AutoProcessor.from_pretrained(model_id)

IMG_URLS = [
    "https://picsum.photos/id/237/400/300",
    "https://picsum.photos/id/231/200/300",
    "https://picsum.photos/id/27/500/500",
    "https://picsum.photos/id/17/150/600",
]
PROMPT = "<s>[INST]Describe the images.\n[IMG][IMG][IMG][IMG][/INST]"

inputs = processor(images=IMG_URLS, text=PROMPT, return_tensors="pt")
generate_ids = model.generate(input_ids=inputs["input_ids"], pixel_values=inputs["pixel_values"][0], attention_mask=inputs["attention_mask"], max_new_tokens=500)
ouptut = processor.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]

config_update9a450cb2

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

Ready to merge

This branch is ready to get merged automatically.

· Sign up or log in to comment