Instructions to use cyankiwi/Trinity-Mini-AWQ-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use cyankiwi/Trinity-Mini-AWQ-4bit with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="cyankiwi/Trinity-Mini-AWQ-4bit", trust_remote_code=True) messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("cyankiwi/Trinity-Mini-AWQ-4bit", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("cyankiwi/Trinity-Mini-AWQ-4bit", trust_remote_code=True, 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=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use cyankiwi/Trinity-Mini-AWQ-4bit with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "cyankiwi/Trinity-Mini-AWQ-4bit" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "cyankiwi/Trinity-Mini-AWQ-4bit", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/cyankiwi/Trinity-Mini-AWQ-4bit
- SGLang
How to use cyankiwi/Trinity-Mini-AWQ-4bit 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 "cyankiwi/Trinity-Mini-AWQ-4bit" \ --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": "cyankiwi/Trinity-Mini-AWQ-4bit", "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 "cyankiwi/Trinity-Mini-AWQ-4bit" \ --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": "cyankiwi/Trinity-Mini-AWQ-4bit", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use cyankiwi/Trinity-Mini-AWQ-4bit with Docker Model Runner:
docker model run hf.co/cyankiwi/Trinity-Mini-AWQ-4bit
Regenerate model card with update_model_card.py (cyankiwi header, GCS banners)
Browse files
README.md
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
| 3 |
language:
|
| 4 |
- en
|
| 5 |
- es
|
|
@@ -13,9 +14,63 @@ language:
|
|
| 13 |
- ko
|
| 14 |
- zh
|
| 15 |
library_name: transformers
|
| 16 |
-
|
| 17 |
-
|
|
|
|
| 18 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
<div align="center">
|
| 20 |
<picture>
|
| 21 |
<img
|
|
@@ -51,7 +106,7 @@ Try it out now at [chat.arcee.ai](http://chat.arcee.ai/)
|
|
| 51 |
* **Experts:** 128 total, 8 active, 1 shared
|
| 52 |
* **Context length:** 128k
|
| 53 |
* **Training Tokens:** 10T
|
| 54 |
-
* **License:** [
|
| 55 |
* **Recommended settings:**
|
| 56 |
* temperature: 0.15
|
| 57 |
* top_k: 50
|
|
@@ -153,7 +208,7 @@ pip install "vllm>=0.11.1"
|
|
| 153 |
Serving the model with suggested settings:
|
| 154 |
|
| 155 |
```
|
| 156 |
-
vllm serve arcee-
|
| 157 |
--dtype bfloat16 \
|
| 158 |
--enable-auto-tool-choice \
|
| 159 |
--reasoning-parser deepseek_r1 \
|
|
@@ -210,4 +265,4 @@ curl -X POST "https://openrouter.ai/v1/chat/completions" \
|
|
| 210 |
|
| 211 |
## License
|
| 212 |
|
| 213 |
-
Trinity-Mini is released under the
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model: arcee-ai/Trinity-Mini
|
| 3 |
+
base_model_relation: quantized
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
- es
|
|
|
|
| 14 |
- ko
|
| 15 |
- zh
|
| 16 |
library_name: transformers
|
| 17 |
+
license: other
|
| 18 |
+
license_link: LICENSE
|
| 19 |
+
license_name: openmdw-1.1
|
| 20 |
---
|
| 21 |
+
|
| 22 |
+
<div align="left">
|
| 23 |
+
<a href="https://cyan.kiwi" target="_blank"><img src="https://storage.googleapis.com/cyankiwi-assets/model-card/model-card-banners-cyankiwi.png" width="800" alt="cyankiwi, the LLM inference agent for open models"></a>
|
| 24 |
+
</div>
|
| 25 |
+
|
| 26 |
+
<div align="left">
|
| 27 |
+
<table style="border-collapse:collapse; border:none;">
|
| 28 |
+
<tr style="border:none;">
|
| 29 |
+
<td align="left" style="border:none; padding:4px 24px 4px 0;"><b>Website</b></td>
|
| 30 |
+
<td align="left" style="border:none; padding:4px 0;"><a href="https://cyan.kiwi/" target="_blank">cyan.kiwi</a></td>
|
| 31 |
+
</tr>
|
| 32 |
+
<tr style="border:none;">
|
| 33 |
+
<td align="left" style="border:none; padding:4px 24px 4px 0;"><b>Models</b></td>
|
| 34 |
+
<td align="left" style="border:none; padding:4px 0;"><a href="https://huggingface.co/cyankiwi" target="_blank">Hugging Face</a> 路 <a href="https://www.modelscope.ai/organization/cyankiwi" target="_blank">ModelScope</a></td>
|
| 35 |
+
</tr>
|
| 36 |
+
<tr style="border:none;">
|
| 37 |
+
<td align="left" style="border:none; padding:4px 24px 4px 0;"><b>Connect</b></td>
|
| 38 |
+
<td align="left" style="border:none; padding:4px 0;"><a href="https://x.com/cyankiwi_ai" target="_blank">X</a> 路 <a href="https://www.reddit.com/r/cyankiwi/" target="_blank">Reddit</a> 路 <a href="mailto:ton@cyan.kiwi">Email</a></td>
|
| 39 |
+
</tr>
|
| 40 |
+
</table>
|
| 41 |
+
</div>
|
| 42 |
+
|
| 43 |
+
<br>
|
| 44 |
+
|
| 45 |
+
<div align="left">
|
| 46 |
+
<img src="https://storage.googleapis.com/cyankiwi-assets/model-card/model-card-banners-awq.png" width="800" alt="cyankiwi Activation-aware Weight Quantization">
|
| 47 |
+
</div>
|
| 48 |
+
|
| 49 |
+
<div align="left">
|
| 50 |
+
<table style="border-collapse:collapse; border:none;">
|
| 51 |
+
<tr style="border:none;">
|
| 52 |
+
<td align="left" style="border:none; padding:4px 24px 4px 0;"><b>Version</b></td>
|
| 53 |
+
<td align="left" style="border:none; padding:4px 0;"><a href="https://github.com/vllm-project/llm-compressor" target="_blank">llm-compressor</a></td>
|
| 54 |
+
</tr>
|
| 55 |
+
<tr style="border:none;">
|
| 56 |
+
<td align="left" style="border:none; padding:4px 24px 4px 0;"><b>Calibration</b></td>
|
| 57 |
+
<td align="left" style="border:none; padding:4px 0;"><a href="https://huggingface.co/datasets/nvidia/Llama-Nemotron-Post-Training-Dataset" target="_blank">nvidia/Llama-Nemotron-Post-Training-Dataset</a></td>
|
| 58 |
+
</tr>
|
| 59 |
+
<tr style="border:none;">
|
| 60 |
+
<td align="left" style="border:none; padding:4px 24px 4px 0;"><b>Model Size</b></td>
|
| 61 |
+
<td align="left" style="border:none; padding:4px 0;">16.35 GB</td>
|
| 62 |
+
</tr>
|
| 63 |
+
</table>
|
| 64 |
+
</div>
|
| 65 |
+
|
| 66 |
+
<br>
|
| 67 |
+
|
| 68 |
+
<p align="left">
|
| 69 |
+
<a href="https://cyan.kiwi/workspace?prompt=Deploy%20cyankiwi/Trinity-Mini-AWQ-4bit" target="_blank"><img src="https://storage.googleapis.com/cyankiwi-assets/model-card/model-card-button-deploy.png" width="333" alt="Deploy on cyan.kiwi private endpoints"></a>
|
| 70 |
+
</p>
|
| 71 |
+
|
| 72 |
+
---
|
| 73 |
+
|
| 74 |
<div align="center">
|
| 75 |
<picture>
|
| 76 |
<img
|
|
|
|
| 106 |
* **Experts:** 128 total, 8 active, 1 shared
|
| 107 |
* **Context length:** 128k
|
| 108 |
* **Training Tokens:** 10T
|
| 109 |
+
* **License:** [OpenMDW-1.1](https://huggingface.co/arcee-ai/Trinity-Mini#license)
|
| 110 |
* **Recommended settings:**
|
| 111 |
* temperature: 0.15
|
| 112 |
* top_k: 50
|
|
|
|
| 208 |
Serving the model with suggested settings:
|
| 209 |
|
| 210 |
```
|
| 211 |
+
vllm serve arcee-ai/Trinity-Mini \
|
| 212 |
--dtype bfloat16 \
|
| 213 |
--enable-auto-tool-choice \
|
| 214 |
--reasoning-parser deepseek_r1 \
|
|
|
|
| 265 |
|
| 266 |
## License
|
| 267 |
|
| 268 |
+
Trinity-Mini is released under the OpenMDW-1.1 license.
|