Instructions to use mykor/Tri-7B-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mykor/Tri-7B-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="mykor/Tri-7B-gguf",
	filename="Tri-7B-BF16.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use mykor/Tri-7B-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mykor/Tri-7B-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mykor/Tri-7B-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf mykor/Tri-7B-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf mykor/Tri-7B-gguf:Q4_K_M

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 mykor/Tri-7B-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf mykor/Tri-7B-gguf:Q4_K_M

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 mykor/Tri-7B-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf mykor/Tri-7B-gguf:Q4_K_M

Use Docker

docker model run hf.co/mykor/Tri-7B-gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use mykor/Tri-7B-gguf with Ollama:
```
ollama run hf.co/mykor/Tri-7B-gguf:Q4_K_M
```

Unsloth Studio new

How to use mykor/Tri-7B-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 mykor/Tri-7B-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 mykor/Tri-7B-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for mykor/Tri-7B-gguf to start chatting

Pi new

How to use mykor/Tri-7B-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf mykor/Tri-7B-gguf:Q4_K_M

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": "mykor/Tri-7B-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use mykor/Tri-7B-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 mykor/Tri-7B-gguf:Q4_K_M

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 mykor/Tri-7B-gguf:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use mykor/Tri-7B-gguf with Docker Model Runner:
```
docker model run hf.co/mykor/Tri-7B-gguf:Q4_K_M
```

Lemonade

How to use mykor/Tri-7B-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull mykor/Tri-7B-gguf:Q4_K_M

Run and chat with the model

lemonade run user.Tri-7B-gguf-Q4_K_M

List all available models

lemonade list

Tri-7B

Introduction

We introduce Tri-7B, the next generation model following Trillion-7B-preview, that continues to push the boundaries of efficient training while achieving exceptional performance at the 7B parameter scale.

Average Performance vs. Approximate Training FLOPs

Key Highlights

Enhanced Reasoning: Modified training dataset mixture specifically optimized for reasoning capabilities
Advanced Post-Training: Significantly improved RL training pipeline focusing on mathematical reasoning and everyday usage
Extended Context: Supports up to 32K context length for long-form understanding
Multi-lingual: Specially optimized for Korean, English, and Japanese.

Our Tri-7B model represents a significant advancement over Trillion-7B-preview, achieving substantial performance improvements across all evaluated domains while maintaining the same efficient parameter count.

Model Specifications

Tri-7B

Type: Causal Language Model
Training Stage: Pre-training & Post-training
Architecture: Transformer Decoder with RoPE, SwiGLU, RMSNorm
Number of Parameters: 7.76B
Number of Layers: 32
Number of Attention Heads: 32
Context Length: 32,768
Vocab Size: 128,256

Quickstart

Here is a code snippet with apply_chat_template that demonstrates how to load the tokenizer and model and generate text.

Tri-7B Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "trillionlabs/Tri-7B"

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

prompt = "Explain the concept of quantum computing in simple terms."
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=512
)
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)

Tri-7B is also available with vLLM and SGLang!

# vLLM
vllm serve trillionlabs/Tri-7B --dtype bfloat16 --max-model-len 32768

# vLLM with custom options
vllm serve trillionlabs/Tri-7B \
    --dtype bfloat16 \
    --max-model-len 32768 \
    --gpu-memory-utilization 0.95 \
    --port 8000

# SGLang
python3 -m sglang.launch_server --model-path trillionlabs/Tri-7B --dtype bfloat16

# SGLang with custom options
python3 -m sglang.launch_server \
    --model-path trillionlabs/Tri-7B \
    --dtype bfloat16 \
    --context-length 32768 \
    --port 30000 \
    --host 0.0.0.0

Evaluation

We evaluated Tri-7B across a comprehensive suite of benchmarks assessing general reasoning, knowledge recall, coding abilities, mathematical reasoning, and instruction-following capabilities. Compared to our previous generation model Trillion-7B-preview, Tri-7B achieves significant gains across all domains.

Full evaluation settings

Benchmark	Language	Evaluation Setting	Metric
General Reasoning and Factuality
• HellaSwag	English	0-shot	accuracy
• ARC:C	English	0-shot	accuracy
• HAERAE	Korean	3-shot	accuracy
• CLIcK	Korean	0-shot	accuracy
• KoBEST	Korean	5-shot	accuracy
Knowledge and Reasoning
• KMMLU	Korean	5-shot	accuracy
• MMLU	English	5-shot	accuracy
• Global-MMLU-Lite-ja	English	5-shot	accuracy
Coding
• HumanEval	English	0-shot	pass@1
• MBPPPlus	English	0-shot	pass@1
Mathematical Reasoning
• GSM8k	English	0-shot, CoT	exact-match
• MATH	English	0-shot, CoT	exact-match
• GPQA	English	4-shot	accuracy
• HRM8k	Korean	0-shot, CoT	exact-match
Instruction Following and Chat
• IFEval	English	0-shot	strict-average
• koIFEval	Korean	0-shot	strict-average
• MT-Bench	English	LLM-as-a-judge (gpt-4o)	LLM score
• KO-MT-Bench	Korean	LLM-as-a-judge (gpt-4o)	LLM score
• systemIFEval	English	0-shot	strict-average

*Note that koIFEval, systemIFEval, and KoRuler are our in-house evaluation benchmarks adapted for Korean to better assess model capabilities in Korean language tasks.
**Note that MT-Bench, KO-MT-Bench, and LogicKor use a 10-point scale.

Benchmark Results

Models compared:

Tri-7B (Next Generation)
Trillion-7B-preview (Previous Generation)

General Reasoning and Factuality

Benchmark	Tri-7B	Trillion-7B-preview	Improvement
HellaSwag	59.52	58.94	+0.58
ARC:C	58.28	54.44	+3.84
HAERAE	82.49	80.02	+2.47
KoBEST	82.72	79.61	+3.11
CLIcK	64.43	60.41	+4.02
KMMLU	51.74	48.09	+3.65
MMLU	68.16	63.52	+4.64
Global-MMLU-Lite-ja	59.25	60.75	-1.50

Coding

Benchmark	Tri-7B	Trillion-7B-preview	Improvement
HumanEval	53.66	55.48	-1.82
MBPPPlus	64.29	58.99	+5.30

Mathematical Reasoning

Benchmark	Tri-7B	Trillion-7B-preview	Improvement
GSM8k	77.94	72.25	+5.69
MATH	49.40	32.70	+16.70
GPQA	34.15	32.81	+1.34
HRM8k	39.08	30.10	+8.98

Instruction Following and Chat

Benchmark	Tri-7B	Trillion-7B-preview	Improvement
IFEval	79.26	79.13	+0.13
koIFEval	76.63	66.58	+10.05
MT-Bench	7.82	6.53	+1.29
KO-MT-Bench	7.64	6.27	+1.37
systemIFEval	66.43	27.28	+39.15

Limitations

Language Support: The model is optimized for English, Korean, and Japanese. Usage with other languages may result in degraded performance.
Knowledge Cutoff: The model's information is limited to data available up to Febuary, 2025.

License

This model is licensed under the Apache License 2.0.

Contact

For inquiries, please contact: info@trillionlabs.co

Downloads last month: 195

GGUF

Model size

8B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for mykor/Tri-7B-gguf

Base model

trillionlabs/Tri-7B

Quantized

(3)

this model