Instructions to use Thunderbolts123/UltraThinker-Coder-3B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Thunderbolts123/UltraThinker-Coder-3B with Transformers:

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

pipe = pipeline("text-generation", model="Thunderbolts123/UltraThinker-Coder-3B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Thunderbolts123/UltraThinker-Coder-3B")
model = AutoModelForMultimodalLM.from_pretrained("Thunderbolts123/UltraThinker-Coder-3B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Thunderbolts123/UltraThinker-Coder-3B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Thunderbolts123/UltraThinker-Coder-3B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Thunderbolts123/UltraThinker-Coder-3B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Thunderbolts123/UltraThinker-Coder-3B

SGLang

How to use Thunderbolts123/UltraThinker-Coder-3B 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 "Thunderbolts123/UltraThinker-Coder-3B" \
    --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": "Thunderbolts123/UltraThinker-Coder-3B",
		"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 "Thunderbolts123/UltraThinker-Coder-3B" \
        --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": "Thunderbolts123/UltraThinker-Coder-3B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Studio

How to use Thunderbolts123/UltraThinker-Coder-3B with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Thunderbolts123/UltraThinker-Coder-3B to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Thunderbolts123/UltraThinker-Coder-3B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Thunderbolts123/UltraThinker-Coder-3B to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="Thunderbolts123/UltraThinker-Coder-3B",
    max_seq_length=2048,
)

Docker Model Runner
How to use Thunderbolts123/UltraThinker-Coder-3B with Docker Model Runner:
```
docker model run hf.co/Thunderbolts123/UltraThinker-Coder-3B
```

Thunderbolts123 commited on 8 days ago

Commit

2863153

verified ·

1 Parent(s): 54a2646

Training in progress, step 200

Browse files

Files changed (3) hide show

README.md +4 -4
adapter_config.json +4 -4
adapter_model.safetensors +1 -1

README.md CHANGED Viewed

@@ -4,9 +4,9 @@ library_name: transformers
 model_name: UltraThinker-Coder-3B
 tags:
 - generated_from_trainer
-- sft
-- trl
 - unsloth
 licence: license
 ---
@@ -37,8 +37,8 @@ This model was trained with SFT.
 - TRL: 0.24.0
 - Transformers: 5.5.0
-- Pytorch: 2.10.0
-- Datasets: 3.6.0
 - Tokenizers: 0.22.2
 ## Citations

 model_name: UltraThinker-Coder-3B
 tags:
 - generated_from_trainer
 - unsloth
+- trl
+- sft
 licence: license
 ---
 - TRL: 0.24.0
 - Transformers: 5.5.0
+- Pytorch: 2.11.0+cu128
+- Datasets: 4.3.0
 - Tokenizers: 0.22.2
 ## Citations

adapter_config.json CHANGED Viewed

@@ -34,13 +34,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "down_proj",
-    "up_proj",
     "k_proj",
     "v_proj",
     "gate_proj",
-    "q_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

   "rank_pattern": {},
   "revision": null,
   "target_modules": [
     "k_proj",
+    "q_proj",
     "v_proj",
     "gate_proj",
+    "up_proj",
+    "down_proj",
+    "o_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4f09b7bfce71373cdc9dbe8b2fab96e5cecbf0354c7d8e7b9e162d8b8120368
 size 479005064

 version https://git-lfs.github.com/spec/v1
+oid sha256:b6772c13288995df15e98b0a9ab858f655b2274ed3d3ace5fc52772cbc4b2689
 size 479005064