StepFun

About StepFun

阶跃星辰于2023年4月成立,以“智能阶跃,十倍每个人的可能”为使命。阶跃星辰坚定自研超级模型,积极布局算力、数据等关键资源,发挥算法和人才优势,已完成 Step-1 千亿参数语言大模型和 Step-1V 千亿多模态大模型的研发,在图像理解、多轮指令跟随、数学能力、逻辑推理、文本创作等方面性能达到业界领先水平。

Website ↗

Source ↗ · 2026-11-12T04:15:32.067864459Z

32 indexed models

32 matching models

stepaudio-3-asr-max

stepFun/stepaudio-3-asr-max

Input modalities
audio
Output modalities
transcription
Capabilities
audioInput, streaming, transcription

StepAudio 3 Gen

stepFun/stepaudio-3-gen

StepAudio 3 Gen creates comprehensive audio content from natural language. Describe roles, timbre, and emotion, and it generates expressive, controllable voice—then adds sound effects, ambient sound, background music, and singing, orchestrated into a single finished clip. Technic

Input modalities
text, audio
Output modalities
audio
Capabilities
audioInput, audioOutput, multimodal, musicGeneration, speechOutput

StepAudio 3 Music

stepFun/stepaudio-3-music

From a natural-language description, lyrics, vocals, or reference audio, the model covers the full arc of music creation—from a rough idea to a finished song to intelligent scoring. Technical report · View more demos · Voice Studio

Input modalities
text, audio
Output modalities
audio
Capabilities
audioInput, audioOutput, multimodal, musicGeneration

StepAudio 3 Realtime

stepFun/stepaudio-3-realtime

A full-duplex voice flagship for realtime interaction. It combines realtime audio understanding, natural interruption handling, turn coordination, adaptive reasoning, think-while-speaking behavior, and Voice Agent tool calling for in-vehicle systems, smart devices, customer servi

Input modalities
audio, text
Output modalities
speech, text
Capabilities
audioInput, multimodal, speechOutput

StepAudio 3 TTS

stepFun/stepaudio-3-tts

The model closely follows natural human speech in timbre, intonation, rhythm, breathing, and pauses. It can express laughter, hesitation, stutters, repetition, and self-correction while adapting emotion and tone to the meaning of the text.

Input modalities
text
Output modalities
speech
Capabilities
speechOutput, streaming

stepfun-ai/GELab-Zero-4B-preview

azureFoundry/HuggingFace/stepfun-ai-gelab-zero-4b-preview@1

`stepfun-ai/GELab-Zero-4B-preview` powered by vLLM - [Original Model Card](https://huggingface.co/stepfun-ai/GELab-Zero-4B-preview) - [vLLM Documentation](https://docs.vllm.ai/) ## Chat Completions API ### Send Request You can use cURL or any REST Client to send a request to

Input modalities
text
Output modalities
text

stepfun-ai/PaCoRe-8B

vertexAI/publishers/hf-stepfun-ai/models/pacore-8b@001

Input modalities
text
Output modalities
text

stepfun-ai/Step3-VL-10B

vertexAI/publishers/hf-stepfun-ai/models/step3-vl-10b@001

Input modalities
image, text
Output modalities
text
Capabilities
multimodal, visionInput

StepFun: step3

siliconFlow/stepfun-ai/step3

Step3 is a cutting-edge multimodal reasoning model from StepFun. It is built on a Mixture-of-Experts (MoE) architecture with 321B total parameters and 38B active parameters. The model is designed end-to-end to minimize decoding costs while delivering top-tier performance in visio

Maximum output
65,536
Input modalities
text, image
Output modalities
text
Capabilities
multimodal, structuredOutputs, toolCalling, visionInput

step-tts-2

stepFun/step-tts-2

Input modalities
text
Output modalities
speech
Capabilities
speechOutput

Step TTS Mini

stepFun/step-tts-mini

Supports Mandarin Chinese, English, Japanese, Cantonese, and Sichuanese with 19 official voices. It also supports high-quality voice cloning in Chinese, English, and Japanese for customer outreach, companion applications, and voice assistants that require natural speech.

Input modalities
text
Output modalities
speech
Capabilities
speechOutput

Models

Loading
EVIDENCE / COMPARISON

Benchmarks

Explore every benchmark

Loading

These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

Install TensorTrace