Instructions to use eeshaAI/zeeb with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use eeshaAI/zeeb with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("EeshaAI/zeeb")
model = PeftModel.from_pretrained(base_model, "eeshaAI/zeeb")

Transformers

How to use eeshaAI/zeeb with Transformers:

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

pipe = pipeline("text-generation", model="eeshaAI/zeeb")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("eeshaAI/zeeb")
model = AutoModelForCausalLM.from_pretrained("eeshaAI/zeeb")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use eeshaAI/zeeb with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "eeshaAI/zeeb"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "eeshaAI/zeeb",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/eeshaAI/zeeb

SGLang

How to use eeshaAI/zeeb 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 "eeshaAI/zeeb" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "eeshaAI/zeeb",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "eeshaAI/zeeb" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "eeshaAI/zeeb",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use eeshaAI/zeeb with Docker Model Runner:
```
docker model run hf.co/eeshaAI/zeeb
```

eeshaAI commited on 18 days ago

Commit

04b5ef6

verified ·

1 Parent(s): e2a1d88

End of training

Browse files

Files changed (1) hide show

README.md +59 -19

README.md CHANGED Viewed

@@ -1,28 +1,68 @@
 ---
 license: apache-2.0
 tags:
-- video-generation
-- visual-tokenization
-- olmo
-- cpu
-- gguf
 ---
-# zeeb — Video-LLM
-A video generation model built by teaching OLMo 2 1B Instruct to "speak" video as a language of visual tokens.
-## Architecture
-- **LLM**: OLMo 2 1B Instruct (Apache 2.0, GGUF Q4_K_M)
-- **Tokenizer**: Custom VQ-VAE (17M params)
-- **Inference**: CPU-only via llama-cpp-python
-## Pipeline
-```
-Text Prompt → LLM → Visual Tokens → VQ-VAE Decoder → Video
-```
-## Files
-- `vq_vae_final.pt` — Trained VQ-VAE tokenizer
-- `tokenized_dataset.json` — Tokenized training data
-- `train_on_hf_spaces.py` — Full LoRA training script

 ---
+library_name: peft
 license: apache-2.0
+base_model: allenai/OLMo-2-0425-1B-Instruct
 tags:
+- base_model:adapter:allenai/OLMo-2-0425-1B-Instruct
+- lora
+- transformers
+pipeline_tag: text-generation
+model-index:
+- name: zeeb
+  results: []
 ---
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+# zeeb
+This model is a fine-tuned version of [allenai/OLMo-2-0425-1B-Instruct](https://huggingface.co/allenai/OLMo-2-0425-1B-Instruct) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.5331
+## Model description
+More information needed
+## Intended uses & limitations
+More information needed
+## Training and evaluation data
+More information needed
+## Training procedure
+### Training hyperparameters
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 0.05
+- num_epochs: 3
+- mixed_precision_training: Native AMP
+### Training results
+| Training Loss | Epoch  | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 11.5396       | 0.8421 | 500  | 11.5391         |
+| 11.5343       | 1.6838 | 1000 | 11.5341         |
+| 11.5332       | 2.5255 | 1500 | 11.5331         |
+### Framework versions
+- PEFT 0.19.1
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.22.2