Instructions to use lunahr/thea-rp-3b-25r with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lunahr/thea-rp-3b-25r with Transformers:

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

pipe = pipeline("text-generation", model="lunahr/thea-rp-3b-25r")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lunahr/thea-rp-3b-25r")
model = AutoModelForCausalLM.from_pretrained("lunahr/thea-rp-3b-25r", device_map="auto")
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 lunahr/thea-rp-3b-25r with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lunahr/thea-rp-3b-25r"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lunahr/thea-rp-3b-25r",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lunahr/thea-rp-3b-25r

SGLang

How to use lunahr/thea-rp-3b-25r 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 "lunahr/thea-rp-3b-25r" \
    --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": "lunahr/thea-rp-3b-25r",
		"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 "lunahr/thea-rp-3b-25r" \
        --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": "lunahr/thea-rp-3b-25r",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lunahr/thea-rp-3b-25r with Docker Model Runner:
```
docker model run hf.co/lunahr/thea-rp-3b-25r
```

lunahr commited on Dec 15, 2024

Commit

9852a47

verified ·

1 Parent(s): 554a31b

updated usernames

Browse files

Files changed (1) hide show

README.md +9 -9

README.md CHANGED Viewed

@@ -13,7 +13,7 @@ tags:
 base_model: SicariusSicariiStuff/Impish_LLAMA_3B
 datasets:
 - KingNish/reasoning-base-20k
-- piotr25691/thea-name-overrides
 model-index:
 - name: thea-rp-3b-25r
   results:
@@ -30,7 +30,7 @@ model-index:
       value: 65.78
       name: strict accuracy
     source:
-      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
@@ -45,7 +45,7 @@ model-index:
       value: 20.01
       name: normalized accuracy
     source:
-      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
@@ -60,7 +60,7 @@ model-index:
       value: 11.71
       name: exact match
     source:
-      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
@@ -75,7 +75,7 @@ model-index:
       value: 3.24
       name: acc_norm
     source:
-      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
@@ -90,7 +90,7 @@ model-index:
       value: 5.93
       name: acc_norm
     source:
-      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
@@ -107,7 +107,7 @@ model-index:
       value: 22.89
       name: accuracy
     source:
-      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-rp-3b-25r
       name: Open LLM Leaderboard
 ---
@@ -125,7 +125,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 MAX_REASONING_TOKENS = 1024
 MAX_RESPONSE_TOKENS = 512
-model_name = "piotr25691/thea-rp-3b-25r"
 model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")
 tokenizer = AutoTokenizer.from_pretrained(model_name)
@@ -153,7 +153,7 @@ response_output = tokenizer.decode(response_ids[0, response_inputs.input_ids.sha
 print("ANSWER: " + response_output)
 ```
-- **Trained by:** [Piotr Zalewski](https://huggingface.co/piotr25691)
 - **License:** llama3.2
 - **Finetuned from model:** [SicariusSicariiStuff/Impish_LLAMA_3B](https://huggingface.co/SicariusSicariiStuff/Impish_LLAMA_3B)
 - **Dataset used:** [KingNish/reasoning-base-20k](https://huggingface.co/datasets/KingNish/reasoning-base-20k)

 base_model: SicariusSicariiStuff/Impish_LLAMA_3B
 datasets:
 - KingNish/reasoning-base-20k
+- lunahr/thea-name-overrides
 model-index:
 - name: thea-rp-3b-25r
   results:
       value: 65.78
       name: strict accuracy
     source:
+      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
       value: 20.01
       name: normalized accuracy
     source:
+      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
       value: 11.71
       name: exact match
     source:
+      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
       value: 3.24
       name: acc_norm
     source:
+      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
       value: 5.93
       name: acc_norm
     source:
+      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-rp-3b-25r
       name: Open LLM Leaderboard
   - task:
       type: text-generation
       value: 22.89
       name: accuracy
     source:
+      url: https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-rp-3b-25r
       name: Open LLM Leaderboard
 ---
 MAX_REASONING_TOKENS = 1024
 MAX_RESPONSE_TOKENS = 512
+model_name = "lunahr/thea-rp-3b-25r"
 model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 print("ANSWER: " + response_output)
 ```
+- **Trained by:** [Piotr Zalewski](https://huggingface.co/lunahr)
 - **License:** llama3.2
 - **Finetuned from model:** [SicariusSicariiStuff/Impish_LLAMA_3B](https://huggingface.co/SicariusSicariiStuff/Impish_LLAMA_3B)
 - **Dataset used:** [KingNish/reasoning-base-20k](https://huggingface.co/datasets/KingNish/reasoning-base-20k)