vibethinker

• Context Length: 131072

• Model Name: vibethinker

• Languages: en, zh

• Abilities: chat, tools

• Description: VibeThinker is a series of dense reasoning language models developed by WeiboAI. Built on the Qwen2 architecture with a post-training methodology centered on the Spectrum-to-Signal Principle (SSP), VibeThinker demonstrates strong reasoning capabilities in mathematics and coding despite its compact size.

Specifications

Model Spec 1 (pytorch, 1_5 Billion)

• Model Format: pytorch

• Model Size (in billions): 1_5

• Quantizations: none

• Engines: vLLM, Transformers, SGLang

• Model ID: WeiboAI/VibeThinker-1.5B

• Model Hubs: Hugging Face, ModelScope

Execute the following command to launch the model, remember to replace ${quantization} with your chosen quantization method from the options listed above:

xinference launch --model-engine ${engine} --model-name vibethinker --size-in-billions 1_5 --model-format pytorch --quantization ${quantization}

Model Spec 2 (pytorch, 3 Billion)

• Model Format: pytorch

• Model Size (in billions): 3

• Quantizations: none

• Engines: vLLM, Transformers, SGLang

• Model ID: WeiboAI/VibeThinker-3B

• Model Hubs: Hugging Face, ModelScope

Execute the following command to launch the model, remember to replace ${quantization} with your chosen quantization method from the options listed above:

xinference launch --model-engine ${engine} --model-name vibethinker --size-in-billions 3 --model-format pytorch --quantization ${quantization}