Instructions to use KakTakOne/VibeThinker-3B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KakTakOne/VibeThinker-3B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="KakTakOne/VibeThinker-3B-GGUF",
	filename="VibeThinker-3B-Q4_K_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use KakTakOne/VibeThinker-3B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf KakTakOne/VibeThinker-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf KakTakOne/VibeThinker-3B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf KakTakOne/VibeThinker-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf KakTakOne/VibeThinker-3B-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 KakTakOne/VibeThinker-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf KakTakOne/VibeThinker-3B-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 KakTakOne/VibeThinker-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf KakTakOne/VibeThinker-3B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/KakTakOne/VibeThinker-3B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use KakTakOne/VibeThinker-3B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KakTakOne/VibeThinker-3B-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": "KakTakOne/VibeThinker-3B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/KakTakOne/VibeThinker-3B-GGUF:Q4_K_M

Ollama
How to use KakTakOne/VibeThinker-3B-GGUF with Ollama:
```
ollama run hf.co/KakTakOne/VibeThinker-3B-GGUF:Q4_K_M
```

Unsloth Studio

How to use KakTakOne/VibeThinker-3B-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 KakTakOne/VibeThinker-3B-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 KakTakOne/VibeThinker-3B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

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

How to use KakTakOne/VibeThinker-3B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf KakTakOne/VibeThinker-3B-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": "KakTakOne/VibeThinker-3B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use KakTakOne/VibeThinker-3B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf KakTakOne/VibeThinker-3B-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 KakTakOne/VibeThinker-3B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use KakTakOne/VibeThinker-3B-GGUF with Docker Model Runner:
```
docker model run hf.co/KakTakOne/VibeThinker-3B-GGUF:Q4_K_M
```

Lemonade

How to use KakTakOne/VibeThinker-3B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull KakTakOne/VibeThinker-3B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.VibeThinker-3B-GGUF-Q4_K_M

List all available models

lemonade list

KakTakOne commited on 11 days ago

Commit

b211bf8

verified ·

1 Parent(s): b15ecac

Upload folder using huggingface_hub

Browse files

Files changed (1) hide show

README.md +137 -162

README.md CHANGED Viewed

@@ -1,162 +1,137 @@
----
-license: mit
-language:
-- en
-- ru
-base_model: WeiboAI/VibeThinker-3B
-tags:
-- math
-- code
-- reasoning
-- gpqa
-- instruction-following
-- gguf
-- llama.cpp
-pipeline_tag: text-generation
----
-# KakTakOne/VibeThinker-3B-GGUF
-This repository contains GGUF format model files for [WeiboAI/VibeThinker-3B](https://huggingface.co/WeiboAI/VibeThinker-3B).
-VibeThinker-3B is a 3-billion-parameter dense reasoning model designed for verifiable reasoning tasks like mathematics, competitive programming, and STEM.
-<details>
-<summary><b>Читать описание на русском языке (Russian Description)</b></summary>
-# KakTakOne/VibeThinker-3B-GGUF
-В этом репозитории содержатся файлы моделей в формате GGUF для [WeiboAI/VibeThinker-3B](https://huggingface.co/WeiboAI/VibeThinker-3B).
-VibeThinker-3B — это модель рассуждений (reasoning model) с 3 миллиардами параметров, сфокусированная на сложных задачах рассуждения с проверяемыми результатами, таких как математика, программирование и STEM.
-## Доступные кванты
-| Имя файла | Тип кванта | Размер файла | Ссылка |
-| --- | --- | --- | --- |
-| [VibeThinker-3B-f16.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-f16.gguf) | FP16 | 6.18 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-f16.gguf?download=true) |
-| [VibeThinker-3B-Q8_0.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q8_0.gguf) | Q8_0 | 3.29 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q8_0.gguf?download=true) |
-| [VibeThinker-3B-Q5_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q5_K_M.gguf) | Q5_K_M | 2.22 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q5_K_M.gguf?download=true) |
-| [VibeThinker-3B-Q4_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q4_K_M.gguf) | Q4_K_M | 1.93 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q4_K_M.gguf?download=true) |
----
-## Введение
-VibeThinker-3B продолжает развитие серии моделей рассуждения VibeThinker на масштабе 3 миллиардов параметров. Благодаря оптимизации пайплайна обучения Spectrum-to-Signal Principle (SSP), модель демонстрирует выдающиеся результаты на бенчмарках AIME, HMMT, IMO-AnswerBench, LiveCodeBench и недавних контестах LeetCode, приближаясь по качеству к флагманским коммерческим моделям рассуждения вроде Qwen3.6 Plus, Gemini 3 Pro, GLM-5 и Kimi K2.5.
-## Ключевые показатели производительности
-* 📏 Модель набирает **76.4** на сложном бенчмарке IMO-AnswerBench (400 олимпиадных задач уровня IMO) с использованием всего 3 млрд параметров, и улучшает результат до **80.6** с применением CLR (Claim-Level Reliability Assessment) на этапе инференса. Это сопоставимо с показателями гораздо более крупных моделей, таких как DeepSeek V3.2 (78.3, 671B), GLM-5 (82.5, 744B) и Kimi K2.5 (81.8, 1T).
-* 🏆 На еженедельных и двухнедельных соревнованиях LeetCode (Python) за период с 25 апреля по 31 мая 2026 года модель успешно прошла **123 из 128** тестов с первой попытки (доля успешных решений составляет **96.1%**).
-## Пайплайн обучения
-Обучение VibeThinker-3B основано на методологии **Spectrum-to-Signal Principle (SSP)**:
-1. **Curriculum SFT в два этапа**: сначала общая кодовая и математическая база, затем сложные рассуждения с длинным контекстом.
-2. **Multi-domain RL** с алгоритмом MaxEnt-Guided Policy Optimization (MGPO) в окне контекста 64K.
-3. **Офлайн дистилляция на себя (Self-Distillation)** для отбора лучших траекторий рассуждений.
-4. **Instruct RL** для улучшения управляемости и форматирования ответов под пользователя.
----
-## Как использовать
-Эти файлы GGUF можно запускать в **LM Studio**, **Ollama**, **llama.cpp** и других совместимых клиентах.
-### LM Studio
-Просто вбей в строку поиска `KakTakOne/VibeThinker-3B-GGUF` и скачай нужный квант.
-### Запуск через консоль (llama.cpp)
-```bash
-llama-cli -m VibeThinker-3B-Q4_K_M.gguf -p "2+2=" -n 128
-```
-</details>
----
-## Available Quantizations
-| File Name | Quant Type | File Size | File Link |
-| --- | --- | --- | --- |
-| [VibeThinker-3B-f16.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-f16.gguf) | FP16 | 6.18 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-f16.gguf?download=true) |
-| [VibeThinker-3B-Q8_0.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q8_0.gguf) | Q8_0 | 3.29 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q8_0.gguf?download=true) |
-| [VibeThinker-3B-Q5_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q5_K_M.gguf) | Q5_K_M | 2.22 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q5_K_M.gguf?download=true) |
-| [VibeThinker-3B-Q4_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q4_K_M.gguf) | Q4_K_M | 1.93 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q4_K_M.gguf?download=true) |
----
-## Introduction
-VibeThinker-3B is a further exploration of the VibeThinker series at the 3B-parameter scale, focusing on challenging reasoning tasks with clear verification signals, such as mathematics, coding, and STEM. By systematically optimizing the Spectrum-to-Signal Principle (SSP) post-training pipeline introduced in VibeThinker-1.5B, VibeThinker-3B achieves strong performance on AIME, HMMT, IMO-AnswerBench, LiveCodeBench and recent LeetCode contests, reaching the performance range of top-tier frontier reasoning models, including Qwen3.6 Plus, Gemini 3 Pro, GLM-5, and Kimi K2.5, on verifiable reasoning benchmarks.
-## Key Performance Data
-* 📏 In terms of reasoning accuracy relative to model scale, VibeThinker-3B reaches **76.4** on IMO-AnswerBench, a highly challenging benchmark with 400 IMO-level problems, with only 3B parameters, and improves to **80.6** with Claim-Level Reliability Assessment (CLR), a test-time scaling strategy. This demonstrates that a model within a strictly small-model regime can reach the performance range of substantially larger models, such as DeepSeek V3.2 (78.3, 671B), GLM-5 (82.5, 744B), and Kimi K2.5 (81.8, 1T).
-* 🏆 To further test the model's out-of-distribution performance, it was evaluated on recent unseen LeetCode weekly and biweekly contests (Python) from Apr. 25 to May 31, 2026. VibeThinker-3B passes **123/128** first-attempt submissions, corresponding to a **96.1%** acceptance rate.
-## Training Pipeline
-VibeThinker-3B follows the **Spectrum-to-Signal Principle (SSP)**. The SFT stage constructs a broad spectrum of valid reasoning trajectories, while the RL stage amplifies correct reasoning signals using verifiable rewards:
-1. **Curriculum-based two-stage SFT** (Stage 1: broad capability coverage, Stage 2: harder/longer samples).
-2. **Multi-domain Reasoning RL** using MaxEnt-Guided Policy Optimization (MGPO) with a 64K context window.
-3. **Offline Self-Distillation** using a learning-potential score to distill high-quality trajectories back into a student model.
-4. **Instruct RL** to improve format controllability on user-facing prompts.
----
-## How to use
-You can load these GGUF files in **LM Studio**, **Ollama**, **llama.cpp**, or any other GGUF-compatible inference engine.
-### LM Studio
-Search for `KakTakOne/VibeThinker-3B-GGUF` directly in LM Studio search bar and download the desired quantization.
-### CLI (llama.cpp)
-```bash
-llama-cli -m VibeThinker-3B-Q4_K_M.gguf -p "2+2=" -n 128
-```
-## Citations & References
-```bibtex
-@misc{xu2026vibethinker3bexploringfrontierverifiable,
-      title={VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models},
-      author={Sen Xu and Shixi Liu and Wei Wang and Jixin Min and Yingwei Dai and Zhibin Yin and Yirong Chen and Xin Zhou and Junlin Zhang},
-      year={2026},
-      eprint={2606.16140},
-      archivePrefix={arXiv},
-      primaryClass={cs.AI},
-      url={https://arxiv.org/abs/2606.16140},
-}
-```
----
-*Quantized by [KakTakOne](https://huggingface.co/KakTakOne) using `llama-quantize`.*� обучения
-Обучение VibeThinker-3B основано на методологии **Spectrum-to-Signal Principle (SSP)**:
-1. **Curriculum SFT в два этапа**: сначала общая кодовая и математическая база, затем сложные рассуждения с длинным контекстом.
-2. **Multi-domain RL** с алгоритмом MaxEnt-Guided Policy Optimization (MGPO) в окне контекста 64K.
-3. **Офлайн дистилляция на себя (Self-Distillation)** для отбора лучших траекторий рассуждений.
-4. **Instruct RL** для улучшения управляемости и форматирования ответов под пользователя.
----
-## Как использовать
-Эти файлы GGUF можно запускать в **LM Studio**, **Ollama**, **llama.cpp** и других совместимых клиентах.
-### LM Studio
-Просто вбей в строку поиска `KakTakOne/VibeThinker-3B-GGUF` и скачай нужный квант.
-### Запуск через консоль (llama.cpp)
-```bash
-llama-cli -m VibeThinker-3B-Q4_K_M.gguf -p "2+2=" -n 128
-```
-</details>
----
-*Quantized by [KakTakOne](https://huggingface.co/KakTakOne) using `llama-quantize`.*

+---
+license: mit
+language:
+- en
+- ru
+base_model: WeiboAI/VibeThinker-3B
+tags:
+- math
+- code
+- reasoning
+- gpqa
+- instruction-following
+- gguf
+- llama.cpp
+pipeline_tag: text-generation
+---
+# KakTakOne/VibeThinker-3B-GGUF
+This repository contains GGUF format model files for [WeiboAI/VibeThinker-3B](https://huggingface.co/WeiboAI/VibeThinker-3B).
+VibeThinker-3B is a 3-billion-parameter dense reasoning model designed for verifiable reasoning tasks like mathematics, competitive programming, and STEM.
+<details>
+<summary><b>Читать описание на русском языке (Russian Description)</b></summary>
+# KakTakOne/VibeThinker-3B-GGUF
+В этом репозитории содержатся файлы моделей в формате GGUF для [WeiboAI/VibeThinker-3B](https://huggingface.co/WeiboAI/VibeThinker-3B).
+VibeThinker-3B — это модель рассуждений (reasoning model) с 3 миллиардами параметров, сфокусированная на сложных задачах рассуждения с проверяемыми результатами, таких как математика, программирование и STEM.
+## Доступные кванты
+| Имя файла | Тип кванта | Размер файла | Ссылка |
+| --- | --- | --- | --- |
+| [VibeThinker-3B-f16.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-f16.gguf) | FP16 | 6.18 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-f16.gguf?download=true) |
+| [VibeThinker-3B-Q8_0.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q8_0.gguf) | Q8_0 | 3.29 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q8_0.gguf?download=true) |
+| [VibeThinker-3B-Q5_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q5_K_M.gguf) | Q5_K_M | 2.22 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q5_K_M.gguf?download=true) |
+| [VibeThinker-3B-Q4_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q4_K_M.gguf) | Q4_K_M | 1.93 ГБ | [Скачать](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q4_K_M.gguf?download=true) |
+---
+## Введение
+VibeThinker-3B продолжает развитие серии моделей рассуждения VibeThinker на масштабе 3 миллиардов параметров. Благодаря оптимизации пайплайна обучения Spectrum-to-Signal Principle (SSP), модель демонстрирует выдающиеся результаты на бенчмарках AIME, HMMT, IMO-AnswerBench, LiveCodeBench и недавних контестах LeetCode, приближаясь по качеству к флагманским коммерческим моделям рассуждения вроде Qwen3.6 Plus, Gemini 3 Pro, GLM-5 и Kimi K2.5.
+## Ключевые показатели производительности
+* 📏 Модель набирает **76.4** на сложном бенчмарке IMO-AnswerBench (400 олимпиадных задач уровня IMO) с использованием всего 3 млрд параметров, и улучшает результат до **80.6** с применением CLR (Claim-Level Reliability Assessment) на этапе инференса. Это сопоставимо с показателями гораздо более крупных моделей, таких как DeepSeek V3.2 (78.3, 671B), GLM-5 (82.5, 744B) и Kimi K2.5 (81.8, 1T).
+* 🏆 На еженедельных и двухнедельных соревнованиях LeetCode (Python) за период с 25 апреля по 31 мая 2026 года модель успешно прошла **123 из 128** тестов с первой попытки (доля успешных решений составляет **96.1%**).
+## Пайплайн обучения
+Обучение VibeThinker-3B основано на методологии **Spectrum-to-Signal Principle (SSP)**:
+1. **Curriculum SFT в два этапа**: сначала общая кодовая и математическая база, затем сложные рассуждения с длинным контекстом.
+2. **Multi-domain RL** с алгоритмом MaxEnt-Guided Policy Optimization (MGPO) в окне контекста 64K.
+3. **Офлайн дистилляция на себя (Self-Distillation)** для отбора лучших траекторий рассуждений.
+4. **Instruct RL** для улучшения управляемости и форматирования ответов под пользователя.
+---
+## Как использовать
+Эти файлы GGUF можно запускать в **LM Studio**, **Ollama**, **llama.cpp** и других совместимых клиентах.
+### LM Studio
+Просто вбей в строку поиска `KakTakOne/VibeThinker-3B-GGUF` и скачай нужный квант.
+### Запуск через консоль (llama.cpp)
+```bash
+llama-cli -m VibeThinker-3B-Q4_K_M.gguf -p "2+2=" -n 128
+```
+</details>
+---
+## Available Quantizations
+| File Name | Quant Type | File Size | File Link |
+| --- | --- | --- | --- |
+| [VibeThinker-3B-f16.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-f16.gguf) | FP16 | 6.18 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-f16.gguf?download=true) |
+| [VibeThinker-3B-Q8_0.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q8_0.gguf) | Q8_0 | 3.29 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q8_0.gguf?download=true) |
+| [VibeThinker-3B-Q5_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q5_K_M.gguf) | Q5_K_M | 2.22 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q5_K_M.gguf?download=true) |
+| [VibeThinker-3B-Q4_K_M.gguf](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/blob/main/VibeThinker-3B-Q4_K_M.gguf) | Q4_K_M | 1.93 GB | [Download](https://huggingface.co/KakTakOne/VibeThinker-3B-GGUF/resolve/main/VibeThinker-3B-Q4_K_M.gguf?download=true) |
+---
+## Introduction
+VibeThinker-3B is a further exploration of the VibeThinker series at the 3B-parameter scale, focusing on challenging reasoning tasks with clear verification signals, such as mathematics, coding, and STEM. By systematically optimizing the Spectrum-to-Signal Principle (SSP) post-training pipeline introduced in VibeThinker-1.5B, VibeThinker-3B achieves strong performance on AIME, HMMT, IMO-AnswerBench, LiveCodeBench, and recent LeetCode contests, reaching the performance range of top-tier frontier reasoning models, including Qwen3.6 Plus, Gemini 3 Pro, GLM-5, and Kimi K2.5, on verifiable reasoning benchmarks.
+## Key Performance Data
+* 📏 In terms of reasoning accuracy relative to model scale, VibeThinker-3B reaches **76.4** on IMO-AnswerBench, a highly challenging benchmark with 400 IMO-level problems, with only 3B parameters, and improves to **80.6** with Claim-Level Reliability Assessment (CLR), a test-time scaling strategy. This demonstrates that a model within a strictly small-model regime can reach the performance range of substantially larger models, such as DeepSeek V3.2 (78.3, 671B), GLM-5 (82.5, 744B), and Kimi K2.5 (81.8, 1T).
+* 🏆 To further test the model's out-of-distribution performance, it was evaluated on recent unseen LeetCode weekly and biweekly contests (Python) from Apr. 25 to May 31, 2026. VibeThinker-3B passes **123/128** first-attempt submissions, corresponding to a **96.1%** acceptance rate.
+## Training Pipeline
+VibeThinker-3B follows the **Spectrum-to-Signal Principle (SSP)**. The SFT stage constructs a broad spectrum of valid reasoning trajectories, while the RL stage amplifies correct reasoning signals using verifiable rewards:
+1. **Curriculum-based two-stage SFT** (Stage 1: broad capability coverage, Stage 2: harder/longer samples).
+2. **Multi-domain Reasoning RL** using MaxEnt-Guided Policy Optimization (MGPO) with a 64K context window.
+3. **Offline Self-Distillation** using a learning-potential score to distill high-quality trajectories back into a student model.
+4. **Instruct RL** to improve format controllability on user-facing prompts.
+---
+## How to use
+You can load these GGUF files in **LM Studio**, **Ollama**, **llama.cpp**, or any other GGUF-compatible inference engine.
+### LM Studio
+Search for `KakTakOne/VibeThinker-3B-GGUF` directly in LM Studio search bar and download the desired quantization.
+### CLI (llama.cpp)
+```bash
+llama-cli -m VibeThinker-3B-Q4_K_M.gguf -p "2+2=" -n 128
+```
+## Citations & References
+```bibtex
+@misc{xu2026vibethinker3bexploringfrontierverifiable,
+      title={VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models},
+      author={Sen Xu and Shixi Liu and Wei Wang and Jixin Min and Yingwei Dai and Zhibin Yin and Yirong Chen and Xin Zhou and Junlin Zhang},
+      year={2026},
+      eprint={2606.16140},
+      archivePrefix={arXiv},
+      primaryClass={cs.AI},
+      url={https://arxiv.org/abs/2606.16140},
+}
+```
+---
+*Quantized by [KakTakOne](https://huggingface.co/KakTakOne) using `llama-quantize`.*