Instructions to use kai-os/Carnice-27b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kai-os/Carnice-27b with Transformers:

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

pipe = pipeline("text-generation", model="kai-os/Carnice-27b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kai-os/Carnice-27b")
model = AutoModelForCausalLM.from_pretrained("kai-os/Carnice-27b")
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

vLLM

How to use kai-os/Carnice-27b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kai-os/Carnice-27b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kai-os/Carnice-27b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kai-os/Carnice-27b

SGLang

How to use kai-os/Carnice-27b 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 "kai-os/Carnice-27b" \
    --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": "kai-os/Carnice-27b",
		"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 "kai-os/Carnice-27b" \
        --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": "kai-os/Carnice-27b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kai-os/Carnice-27b with Docker Model Runner:
```
docker model run hf.co/kai-os/Carnice-27b
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Carnice-27b

Carnice-27b is the merged full-model release of the Trinity Hermes-Agent training run on top of Qwen/Qwen3.5-27B.

This repo contains the merged Stage C weights, not just the adapter. The adapter was trained in three stages and then merged back into the base model so it can load as a standalone checkpoint.

Acknowledgements

This work would not have been possible without Zachary Mueller, Lambda, Teknium, and Nous Research.

Trained using traces from lambda/hermes-agent-reasoning-traces

Trinity Process

Stage A: Premium Reasoning Backbone

3300 train rows
193 validation rows
12288 max length
final eval loss 0.5316
final eval perplexity 1.7016

Stage B: Hermes Alignment

widened Carnice + DJ + Lambda alignment mix
2269 train rows
80 validation rows
final eval loss 0.2336
final eval perplexity 1.2632

Stage C: Carnice Polish

600 train rows
60 validation rows
final eval loss 0.2310
final eval perplexity 1.2599

Intended Use

Carnice-27b is tuned for Hermes-Agent style terminal, file, browser, repo, debugging, and multi-step tool workflows.

Benchmark Status

Reproducible benchmark runs are not attached yet. They will be added only after the dedicated benchmark box run is complete.

Downloads last month: 47

Safetensors

Model size

27B params

Tensor type

BF16

Model tree for kai-os/Carnice-27b

Base model

Qwen/Qwen3.5-27B

Finetuned

(305)

this model

Quantizations

5 models

Collection including kai-os/Carnice-27b

Carnice-V1

Collection

Carnice is a project to make a model preform well in a specific agentic harness (hermes-agent • 6 items • Updated Apr 25 • 17