> ## Documentation Index
> Fetch the complete documentation index at: https://platform.stepfun.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Model Overview

[Pricing & Rate Limits](https://platform.stepfun.ai/docs/en/guides/pricing/details.md) · [Model Lifecycle & Deprecations](https://platform.stepfun.ai/docs/en/guides/model-migration.md) · [API Quickstart](https://platform.stepfun.ai/docs/en/quickstart/overview.md)

> [tab] All models
>
> All models
>
> ## All public models
>
> Reasoning / multimodal
>
> Step 5 PreviewFlagship model for agentic work
>
> Max context1M
>
> StepFun's flagship model for agentic work. It delivers frontier-level performance across software engineering and professional knowledge work, with particular strength in finance. It natively supports text, image, and video input and can use tools and documents to carry multi-step tasks through to completion. API model ID: step-5-preview.
>
> Knowledge work
> Coding
> Agent
> Image and video understanding
>
> Related links
>
> [QuickstartSend your first request](https://platform.stepfun.ai/docs/en/quickstart/overview.md)
>
> [Model overviewCapabilities, specifications, and configuration](https://platform.stepfun.ai/docs/en/guides/models/step-5-preview.md)
>
> Reasoning / multimodal
>
> Step 3.7 FlashFlagship multimodal reasoning
>
> Max context256K
>
> StepFun's flagship multimodal reasoning model. Building on Step 3.5 Flash's high-speed reasoning and tool-calling capabilities, it adds native multimodal input and can understand images and videos without an additional vision MCP or auxiliary model. It supports low, medium, and high reasoning effort.
>
> Reasoning
> Multimodal
> Agent
> Image understanding
> Video understanding
>
> Related links
>
> [Model pageStep 3.7 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.7-flash.md)
>
> [QuickstartMultimodal quickstart](https://platform.stepfun.ai/docs/en/quickstart/overview.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> Reasoning / text
>
> Step 3.5 FlashFlagship reasoning
>
> Max context256K
>
> A flagship reasoning model built for agents. It combines deep reasoning with fast responses and stable, reliable tool calling, and is particularly effective for complex project planning and long-horizon task execution.
>
> Reasoning
> Tool calling
>
> Related links
>
> [Model pageStep 3.5 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.5-flash.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)
>
> Reasoning / text
>
> Step 3.5 Flash 2603Optimized for agents
>
> Max context256K
>
> Optimized from Step 3.5 Flash for high-frequency agent workloads. It preserves flagship reasoning and tool calling while improving token efficiency and inference speed, adds a low-reasoning mode to reduce consumption, and improves compatibility with coding tools and agent frameworks.
>
> Reasoning
> Agent
> Low-reasoning mode
>
> Related links
>
> [Model pageStep 3.5 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.5-flash.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)
>
> Vision
>
> Step-1o Turbo VisionImage and video understanding
>
> Max context32K
>
> Designed for social content creation, AI assistants, and image and video understanding. A single request accepts up to 60 images with a combined size of up to 20 MB, or an MP4 video smaller than 128 MB.
>
> Image understanding
> Video understanding
> Text generation
>
> Related links
>
> [Model pageVision models](https://platform.stepfun.ai/docs/en/guides/models/vision.md)
>
> [GuideVideo understanding guide](https://platform.stepfun.ai/docs/en/guides/developer/video-chat.md)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)
>
> Realtime voice
>
> StepAudio 3 RealtimeNatural conversation with reasoning and action
>
> InteractionVoice to voice
>
> 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 service, and companion applications.
>
> Full duplex
> Adaptive reasoning
> Voice Agent
>
> Related links
>
> [Model pageStepAudio 3 Realtime](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md)
>
> [API referenceRealtime Chat API](https://platform.stepfun.ai/docs/en/api-reference/realtime/chat.md)
>
> Voice conversation
>
> StepAudio 2.5 ChatNatural, human-like conversation
>
> OutputText only
>
> A conversational model designed for natural, human-like interaction with text output. It understands complex meaning and paralinguistic cues such as hesitation and laughter, produces emotionally aware responses, and supports fine-grained persona customization.
>
> Speech understanding
> Paralinguistic cues
> Text output
>
> Related links
>
> [Model pageStepAudio 2.5 Chat](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-2.5-chat)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)
>
> Speech synthesis
>
> StepAudio 3 TTSHuman-level speaking performance
>
> 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.
>
> Human-level expression
> Low-latency streaming
> Unified audio generation
>
> Related links
>
> [Model pageStepAudio 3 TTS](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-tts.md)
>
> [GuideText-to-speech guide](https://platform.stepfun.ai/docs/en/guides/developer/tts.md)
>
> [API referenceText-to-Speech API](https://platform.stepfun.ai/docs/en/api-reference/audio/create-audio.md)
>
> Speech synthesis
>
> Step TTS MiniExpressive text-to-speech
>
> Max input1,000 chars
>
> 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.
>
> Speech generation
> Voice cloning
> Multilingual
>
> Related links
>
> [Model pageAudio models](https://platform.stepfun.ai/docs/en/guides/models/audio.md)
>
> [GuideText-to-speech guide](https://platform.stepfun.ai/docs/en/guides/developer/tts.md)
>
> [API referenceText-to-Speech API](https://platform.stepfun.ai/docs/en/api-reference/audio/create-audio.md)
>
> Speech recognition
>
> StepAudio 2.5 ASRNext-generation streaming ASR flagship
>
> Model scale4B MTP
>
> A next-generation streaming speech recognition model based on a 4B MTP architecture. It balances recognition accuracy and response latency, supports Chinese and English with ITN text normalization, and is designed for realtime captions, voice input, and meeting transcription.
>
> High accuracy
> Low latency
> Chinese and English
>
> Related links
>
> [Model pageStepAudio 2.5 ASR](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-2.5-asr)
>
> [API referenceSpeech Recognition API](https://platform.stepfun.ai/docs/en/api-reference/audio/asr-sse.md)
>
> Speech recognition
>
> StepAudio 2 ASR Pro32B ASR Pro
>
> Parameters32B
>
> A 32B-parameter professional automatic speech recognition model.
>
> Speech recognition
> Large-scale model
> Professional ASR
>
> Related links
>
> [Model pageSpeech recognition models](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-2.5-asr)
>
> [API referenceSpeech Recognition API](https://platform.stepfun.ai/docs/en/api-reference/audio/asr-sse.md)
>
> Speech recognition
>
> Step ASRRealtime and offline recognition
>
> Max file size100 MB
>
> An ASR model with strong Chinese and English recognition. It distinguishes speech from noise and supports mixed Chinese-English speech and heavily accented Mandarin for voice input, voice control, and meeting transcription.
>
> Speech recognition
> Chinese and English
> Accent support
>
> Related links
>
> [Model pageAudio models](https://platform.stepfun.ai/docs/en/guides/models/audio.md)
>
> [API referenceSpeech Recognition API](https://platform.stepfun.ai/docs/en/api-reference/audio/asr-sse.md)
>
> Realtime voice
>
> Step-1o AudioStable realtime voice
>
> Max interaction30 min
>
> Provides ultra-low-latency, bidirectional voice conversation. It accepts Chinese, English, and heavily accented Mandarin, and can respond in Chinese, English, Japanese, Cantonese, and Sichuanese. A single interaction can last up to 30 minutes and process up to 70 minutes of audio.
>
> Realtime interaction
> Multilingual
> Low latency
>
> Related links
>
> [Model pageRealtime voice models](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md)
>
> [API referenceRealtime Chat API](https://platform.stepfun.ai/docs/en/api-reference/realtime/chat.md)

> [tab] Reasoning
>
> Reasoning
>
> ## Reasoning models
>
> Reasoning / multimodal
>
> Step 5 PreviewFlagship model for agentic work
>
> Max context1M
>
> StepFun's flagship model for agentic work. It delivers frontier-level performance across software engineering and professional knowledge work, with particular strength in finance. It natively supports text, image, and video input and can use tools and documents to carry multi-step tasks through to completion. API model ID: step-5-preview.
>
> Knowledge work
> Coding
> Agent
> Image and video understanding
>
> Related links
>
> [QuickstartSend your first request](https://platform.stepfun.ai/docs/en/quickstart/overview.md)
>
> [Model overviewCapabilities, specifications, and configuration](https://platform.stepfun.ai/docs/en/guides/models/step-5-preview.md)
>
> Reasoning / multimodal
>
> Step 3.7 FlashFlagship multimodal reasoning
>
> Max context256K
>
> StepFun's flagship multimodal reasoning model. Building on Step 3.5 Flash's high-speed reasoning and tool-calling capabilities, it adds native multimodal input and can understand images and videos without an additional vision MCP or auxiliary model. It supports low, medium, and high reasoning effort.
>
> Reasoning
> Multimodal
> Agent
> Image understanding
> Video understanding
>
> Related links
>
> [Model pageStep 3.7 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.7-flash.md)
>
> [QuickstartMultimodal quickstart](https://platform.stepfun.ai/docs/en/quickstart/overview.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> Reasoning / text
>
> Step 3.5 FlashFlagship reasoning
>
> Max context256K
>
> A flagship reasoning model built for agents. It combines deep reasoning with fast responses and stable, reliable tool calling, and is particularly effective for complex project planning and long-horizon task execution.
>
> Reasoning
> Tool calling
>
> Related links
>
> [Model pageStep 3.5 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.5-flash.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)
>
> Reasoning / text
>
> Step 3.5 Flash 2603Optimized for agents
>
> Max context256K
>
> Optimized from Step 3.5 Flash for high-frequency agent workloads. It preserves flagship reasoning and tool calling while improving token efficiency and inference speed, adds a low-reasoning mode to reduce consumption, and improves compatibility with coding tools and agent frameworks.
>
> Reasoning
> Agent
> Low-reasoning mode
>
> Related links
>
> [Model pageStep 3.5 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.5-flash.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)

> [tab] Vision
>
> Vision
>
> ## Vision understanding models
>
> Reasoning / multimodal
>
> Step 5 PreviewFlagship model for agentic work
>
> Max context1M
>
> StepFun's flagship model for agentic work. It delivers frontier-level performance across software engineering and professional knowledge work, with particular strength in finance. It natively supports text, image, and video input and can use tools and documents to carry multi-step tasks through to completion. API model ID: step-5-preview.
>
> Knowledge work
> Coding
> Agent
> Image and video understanding
>
> Related links
>
> [QuickstartSend your first request](https://platform.stepfun.ai/docs/en/quickstart/overview.md)
>
> [Model overviewCapabilities, specifications, and configuration](https://platform.stepfun.ai/docs/en/guides/models/step-5-preview.md)
>
> Reasoning / multimodal
>
> Step 3.7 FlashFlagship multimodal reasoning
>
> Max context256K
>
> StepFun's flagship multimodal reasoning model. Building on Step 3.5 Flash's high-speed reasoning and tool-calling capabilities, it adds native multimodal input and can understand images and videos without an additional vision MCP or auxiliary model. It supports low, medium, and high reasoning effort.
>
> Reasoning
> Multimodal
> Agent
> Image understanding
> Video understanding
>
> Related links
>
> [Model pageStep 3.7 Flash](https://platform.stepfun.ai/docs/en/guides/models/step-3.7-flash.md)
>
> [QuickstartMultimodal quickstart](https://platform.stepfun.ai/docs/en/quickstart/overview.md)
>
> [GuideReasoning model guide](https://platform.stepfun.ai/docs/en/guides/developer/reasoning.md)
>
> Vision
>
> Step-1o Turbo VisionImage and video understanding
>
> Max context32K
>
> Designed for social content creation, AI assistants, and image and video understanding. A single request accepts up to 60 images with a combined size of up to 20 MB, or an MP4 video smaller than 128 MB.
>
> Image understanding
> Video understanding
> Text generation
>
> Related links
>
> [Model pageVision models](https://platform.stepfun.ai/docs/en/guides/models/vision.md)
>
> [GuideVideo understanding guide](https://platform.stepfun.ai/docs/en/guides/developer/video-chat.md)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)

> [tab] Audio
>
> Audio
>
> ## Audio and realtime voice models
>
> Realtime voice
>
> StepAudio 3 RealtimeNatural conversation with reasoning and action
>
> InteractionVoice to voice
>
> 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 service, and companion applications.
>
> Full duplex
> Adaptive reasoning
> Voice Agent
>
> Related links
>
> [Model pageStepAudio 3 Realtime](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md)
>
> [API referenceRealtime Chat API](https://platform.stepfun.ai/docs/en/api-reference/realtime/chat.md)
>
> Voice conversation
>
> StepAudio 2.5 ChatNatural, human-like conversation
>
> OutputText only
>
> A conversational model designed for natural, human-like interaction with text output. It understands complex meaning and paralinguistic cues such as hesitation and laughter, produces emotionally aware responses, and supports fine-grained persona customization.
>
> Speech understanding
> Paralinguistic cues
> Text output
>
> Related links
>
> [Model pageStepAudio 2.5 Chat](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-2.5-chat)
>
> [API referenceChat Completion API](https://platform.stepfun.ai/docs/en/api-reference/chat/chat-completion-create.md)
>
> Speech synthesis
>
> StepAudio 3 TTSHuman-level speaking performance
>
> 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.
>
> Human-level expression
> Low-latency streaming
> Unified audio generation
>
> Related links
>
> [Model pageStepAudio 3 TTS](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-tts.md)
>
> [GuideText-to-speech guide](https://platform.stepfun.ai/docs/en/guides/developer/tts.md)
>
> [API referenceText-to-Speech API](https://platform.stepfun.ai/docs/en/api-reference/audio/create-audio.md)
>
> Speech synthesis
>
> Step TTS MiniExpressive text-to-speech
>
> Max input1,000 chars
>
> 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.
>
> Speech generation
> Voice cloning
> Multilingual
>
> Related links
>
> [Model pageAudio models](https://platform.stepfun.ai/docs/en/guides/models/audio.md)
>
> [GuideText-to-speech guide](https://platform.stepfun.ai/docs/en/guides/developer/tts.md)
>
> [API referenceText-to-Speech API](https://platform.stepfun.ai/docs/en/api-reference/audio/create-audio.md)
>
> Speech recognition
>
> StepAudio 2.5 ASRNext-generation streaming ASR flagship
>
> Model scale4B MTP
>
> A next-generation streaming speech recognition model based on a 4B MTP architecture. It balances recognition accuracy and response latency, supports Chinese and English with ITN text normalization, and is designed for realtime captions, voice input, and meeting transcription.
>
> High accuracy
> Low latency
> Chinese and English
>
> Related links
>
> [Model pageStepAudio 2.5 ASR](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-2.5-asr)
>
> [API referenceSpeech Recognition API](https://platform.stepfun.ai/docs/en/api-reference/audio/asr-sse.md)
>
> Speech recognition
>
> StepAudio 2 ASR Pro32B ASR Pro
>
> Parameters32B
>
> A 32B-parameter professional automatic speech recognition model.
>
> Speech recognition
> Large-scale model
> Professional ASR
>
> Related links
>
> [Model pageSpeech recognition models](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-2.5-asr)
>
> [API referenceSpeech Recognition API](https://platform.stepfun.ai/docs/en/api-reference/audio/asr-sse.md)
>
> Speech recognition
>
> Step ASRRealtime and offline recognition
>
> Max file size100 MB
>
> An ASR model with strong Chinese and English recognition. It distinguishes speech from noise and supports mixed Chinese-English speech and heavily accented Mandarin for voice input, voice control, and meeting transcription.
>
> Speech recognition
> Chinese and English
> Accent support
>
> Related links
>
> [Model pageAudio models](https://platform.stepfun.ai/docs/en/guides/models/audio.md)
>
> [API referenceSpeech Recognition API](https://platform.stepfun.ai/docs/en/api-reference/audio/asr-sse.md)
>
> Realtime voice
>
> Step-1o AudioStable realtime voice
>
> Max interaction30 min
>
> Provides ultra-low-latency, bidirectional voice conversation. It accepts Chinese, English, and heavily accented Mandarin, and can respond in Chinese, English, Japanese, Cantonese, and Sichuanese. A single interaction can last up to 30 minutes and process up to 70 minutes of audio.
>
> Realtime interaction
> Multilingual
> Low latency
>
> Related links
>
> [Model pageRealtime voice models](https://platform.stepfun.ai/docs/en/guides/models/stepaudio-3-realtime.md)
>
> [API referenceRealtime Chat API](https://platform.stepfun.ai/docs/en/api-reference/realtime/chat.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.