Instructions to use KnutJaegersberg/Tri-21B-Think-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KnutJaegersberg/Tri-21B-Think-gguf with Transformers:

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

pipe = pipeline("text-generation", model="KnutJaegersberg/Tri-21B-Think-gguf")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("KnutJaegersberg/Tri-21B-Think-gguf", dtype="auto")

llama-cpp-python

How to use KnutJaegersberg/Tri-21B-Think-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="KnutJaegersberg/Tri-21B-Think-gguf",
	filename="Tri-21B-8bit.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use KnutJaegersberg/Tri-21B-Think-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf KnutJaegersberg/Tri-21B-Think-gguf
# Run inference directly in the terminal:
llama-cli -hf KnutJaegersberg/Tri-21B-Think-gguf

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf KnutJaegersberg/Tri-21B-Think-gguf
# Run inference directly in the terminal:
llama-cli -hf KnutJaegersberg/Tri-21B-Think-gguf

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf KnutJaegersberg/Tri-21B-Think-gguf
# Run inference directly in the terminal:
./llama-cli -hf KnutJaegersberg/Tri-21B-Think-gguf

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf KnutJaegersberg/Tri-21B-Think-gguf
# Run inference directly in the terminal:
./build/bin/llama-cli -hf KnutJaegersberg/Tri-21B-Think-gguf

Use Docker

docker model run hf.co/KnutJaegersberg/Tri-21B-Think-gguf

LM Studio
Jan

vLLM

How to use KnutJaegersberg/Tri-21B-Think-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KnutJaegersberg/Tri-21B-Think-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KnutJaegersberg/Tri-21B-Think-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KnutJaegersberg/Tri-21B-Think-gguf

SGLang

How to use KnutJaegersberg/Tri-21B-Think-gguf 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 "KnutJaegersberg/Tri-21B-Think-gguf" \
    --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": "KnutJaegersberg/Tri-21B-Think-gguf",
		"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 "KnutJaegersberg/Tri-21B-Think-gguf" \
        --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": "KnutJaegersberg/Tri-21B-Think-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use KnutJaegersberg/Tri-21B-Think-gguf with Ollama:
```
ollama run hf.co/KnutJaegersberg/Tri-21B-Think-gguf
```

Unsloth Studio new

How to use KnutJaegersberg/Tri-21B-Think-gguf 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 KnutJaegersberg/Tri-21B-Think-gguf 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 KnutJaegersberg/Tri-21B-Think-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for KnutJaegersberg/Tri-21B-Think-gguf to start chatting

Pi new

How to use KnutJaegersberg/Tri-21B-Think-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf KnutJaegersberg/Tri-21B-Think-gguf

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "KnutJaegersberg/Tri-21B-Think-gguf"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use KnutJaegersberg/Tri-21B-Think-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf KnutJaegersberg/Tri-21B-Think-gguf

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default KnutJaegersberg/Tri-21B-Think-gguf

Run Hermes

hermes

Docker Model Runner
How to use KnutJaegersberg/Tri-21B-Think-gguf with Docker Model Runner:
```
docker model run hf.co/KnutJaegersberg/Tri-21B-Think-gguf
```

Lemonade

How to use KnutJaegersberg/Tri-21B-Think-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull KnutJaegersberg/Tri-21B-Think-gguf

Run and chat with the model

lemonade run user.Tri-21B-Think-gguf-{{QUANT_TAG}}

List all available models

lemonade list

The License for the Base Model Tri-21B is the trillion license in this repo, the think and think-preview versions are Apache 2.0.

Tri-21B-Think-Preview

Introduction

Tri-21B-Think-Preview is an intermediate checkpoint of Tri-21B-Think, featuring mid-training context length expansion to 32K tokens and instruction tuning for chain-of-thought reasoning and tool use.

Model Specifications

Type: Causal Language Model (Reasoning-Enhanced)
Base Model: Tri-21B
Architecture: Transformer Decoder with RoPE, SwiGLU, RMSNorm, and GQA
Number of Parameters: 20.73B
Number of Layers: 40
Number of Attention Heads: 32 (Query) / 8 (Key, Value)
Head Dimension: 160
Hidden Size: 5,120
Intermediate Size: 27,392
Context Length: 32,768 (up to 262,144 with YaRN)
Vocab Size: 124,416

Quickstart

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "trillionlabs/Tri-21B-Think-Preview"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "Solve the following step by step: What is the sum of the first 100 prime numbers?"
messages = [
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=4096,
    temperature=0.6,
    top_p=0.9
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)

vLLM & SGLang Deployment

vLLM and SGLang support for Trillion Model is on the way. Stay tuned!

Fine-tuning Notes

Note on <think> tags: This model was trained without <think> and </think> as special tokens. They were added post-training for compatibility with reasoning parsers. If you plan to fine-tune this model, you'll need to modify tokenizer_config.json to avoid indexing errors.

Replace tokens 123975 and 123976 in tokenizer_config.json:

"123975": {
  "content": "<|reserved_special_token_9|>",
  "lstrip": false,
  "normalized": false,
  "rstrip": false,
  "single_word": false,
  "special": true
},
"123976": {
  "content": "<|reserved_special_token_10|>",
  "lstrip": false,
  "normalized": false,
  "rstrip": false,
  "single_word": false,
  "special": true
}

Evaluation

Category	Benchmark	Description	Tri-21B-Think-Preview
Reasoning	GPQA-Diamond	Graduate-level science questions across physics, chemistry, and biology (PhD-level)	54
	AIME 2025	American Invitational Mathematics Examination 2025	50.0
	MMLU-Pro	Massive Multitask Language Understanding with more answer choices and reasoning-focused questions	65.19
	HLE	Humanity's Last Exam — 2,500 expert-level questions across 100+ subjects created by nearly 1,000 domain experts	5.12
Coding	LiveCodeBench v6	Competitive programming benchmark with problems sourced from recent programming contests	48.57
	SciCode	Code generation across 338 subproblems in 16 natural science fields drawn from real research workflows	18
Instruction Following	IFEval	Tests ability to follow precise formatting and output constraint instructions	84.05
	IFBench	Evaluates generalization to novel, verifiable output constraints not seen during training (Allen AI)	51.02
Agentic	TAU2-Bench (Telecom)	Dual-control conversational benchmark where both agent and user use tools to resolve telecom scenarios (Sierra)	93
	AA-LCR	Long-context reasoning over multiple documents at 10K–100K tokens (Artificial Analysis)	15
	AA-Omniscience	Factual reliability across 6,000 questions in 42 subtopics, penalizing hallucinations (Artificial Analysis)	-48.55
Korean	KMMLU-Pro	2,822 questions from 14 Korean National Professional Licensure exams (LG AI Research)	54.18
	CLIcK	1,995 Korean cultural and linguistic knowledge questions sourced from official exams and textbooks (KAIST)	77.94
	KoBALT	Korean linguistic understanding across syntax, semantics, pragmatics, phonetics, and morphology (SNU)	47.29

Limitations

Language Support: Optimized for English, Korean, and Japanese. Other languages may show degraded performance.
Knowledge Cutoff: February 2025.
Intermediate Checkpoint: See Tri-21B-Think for the final model.

License

This model is licensed under the Apache 2.0 License.

Contact

For inquiries: info@trillionlabs.co

Downloads last month: 257

GGUF

Model size

21B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for KnutJaegersberg/Tri-21B-Think-gguf

Base model

trillionlabs/Tri-21B

Quantized

(2)

this model