Instructions to use TildeAI/TildeOpen-30b-64k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TildeAI/TildeOpen-30b-64k with Transformers:

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

pipe = pipeline("text-generation", model="TildeAI/TildeOpen-30b-64k", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("TildeAI/TildeOpen-30b-64k", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("TildeAI/TildeOpen-30b-64k", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use TildeAI/TildeOpen-30b-64k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TildeAI/TildeOpen-30b-64k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TildeAI/TildeOpen-30b-64k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TildeAI/TildeOpen-30b-64k

SGLang

How to use TildeAI/TildeOpen-30b-64k 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 "TildeAI/TildeOpen-30b-64k" \
    --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": "TildeAI/TildeOpen-30b-64k",
		"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 "TildeAI/TildeOpen-30b-64k" \
        --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": "TildeAI/TildeOpen-30b-64k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TildeAI/TildeOpen-30b-64k with Docker Model Runner:
```
docker model run hf.co/TildeAI/TildeOpen-30b-64k
```

Martins Kronis commited on Apr 17

Commit

e4c6ad4

1 Parent(s): 695086a

update code snippets

Browse files

Files changed (1) hide show

README.md +11 -1

README.md CHANGED Viewed

@@ -93,6 +93,7 @@ We follow [Deespeek v3](https://arxiv.org/pdf/2412.19437) and slightly overscale
 ## Running model using HF ```transformers >= 5```
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 # Load tokenizer + model
@@ -113,6 +114,8 @@ outputs = model.generate(
     repetition_penalty=1.2,
     do_sample=False,
 )
 ```
 ## Running model using (old) HF ```transformers < 5```
@@ -123,6 +126,7 @@ We suggest avoiding patches and using transformers >= 5 or vLLM.
 ```python
 from transformers import AutoTokenizer, AutoModelForCausalLM
 # Load tokenizer + model
@@ -135,7 +139,11 @@ model = AutoModelForCausalLM.from_pretrained(
 )
 # Tokenize
-inputs = tokenizer(user_in, return_tensors="pt").to(model.device)
 # Generate (greedy, deterministic)
 outputs = model.generate(
@@ -144,6 +152,8 @@ outputs = model.generate(
     repetition_penalty=1.2,
     do_sample=False,
 )
 ```
 # Evaluation

 ## Running model using HF ```transformers >= 5```
 ```python
+import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 # Load tokenizer + model
     repetition_penalty=1.2,
     do_sample=False,
 )
+text = tokenizer.decode(outputs[0], skip_special_tokens=True)
 ```
 ## Running model using (old) HF ```transformers < 5```
 ```python
+import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
 # Load tokenizer + model
 )
 # Tokenize
+inputs = tokenizer(
+    user_in,
+    return_tensors="pt",
+    return_token_type_ids=False, # sometimes needed for older transformers
+).to(model.device)
 # Generate (greedy, deterministic)
 outputs = model.generate(
     repetition_penalty=1.2,
     do_sample=False,
 )
+text = tokenizer.decode(outputs[0], skip_special_tokens=True)
 ```
 # Evaluation