SiliconFlow

About SiliconFlow

Lightning-fast AI platform for developers. Deploy, fine-tune, and run 200+ optimized LLMs and multimodal models with simple APIs - SiliconFlow.

Website ↗Documentation ↗

Source ↗ · 2026-11-12T03:22:34.644971125Z

Service status

unknown

Current inference service status unverified

Official status page ↗

144 indexed models

144 matching models

Qwen: Qwen3 Next 80B A3B Instruct

qwen/qwen3-next-80b-a3b-instruct-2509

Qwen3-Next-80B-A3B-Instruct is an instruction-tuned chat model in the Qwen3-Next series optimized for fast, stable responses without “thinking” traces. It targets complex tasks across reasoning, code generation, knowledge QA, and multilingual use, while remaining robust on alignm

Context window
262,144
Input modalities
text
Output modalities
text
Capabilities
structuredOutputs, toolCalling

Benchmarks

  • Artificial Analysis: 9.6
  • OpenCompass LLM: 81.3%
  • OpenRouter Benchmarks: 71.324%

Qwen: Qwen3-Omni-30B-A3B-Captioner

siliconFlow/Qwen/Qwen3-Omni-30B-A3B-Captioner

Qwen3-Omni-30B-A3B-Captioner is a Vision-Language Model (VLM) from Alibaba's Qwen team, part of the Qwen3 series. It is specifically designed for generating high-quality, detailed, and accurate image captions. Based on a 30B total parameter Mixture of Experts (MoE) architecture,

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

Qwen: Qwen3-Omni-30B-A3B-Instruct

siliconFlow/Qwen/Qwen3-Omni-30B-A3B-Instruct

Qwen3-Omni-30B-A3B-Instruct is a member of the latest Qwen3 series from Alibaba's Qwen team. It is a Mixture of Experts (MoE) model with 30 billion total parameters and 3 billion active parameters, which effectively reduces inference costs while maintaining powerful performance.

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

Benchmarks

  • MTEB: 27.902%

Qwen: Qwen3-Omni-30B-A3B-Thinking

siliconFlow/Qwen/Qwen3-Omni-30B-A3B-Thinking

Qwen3-Omni-30B-A3B-Thinking is the core "Thinker" component within the Qwen3-Omni omni-modal model's "Thinker-Talker" architecture. It is specifically designed to process multimodal inputs, including text, audio, images, and video, and to execute complex chain-of-thought reasonin

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

Benchmarks

  • MTEB: 25.866%

Qwen: Qwen3 VL 235B A22B Instruct

qwen/qwen3-vl-235b-a22b-instruct

Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table extraction, multilingual OCR). The seri

Context window
262,144
Input modalities
text, image
Output modalities
text
Capabilities
multimodal, promptCaching, structuredOutputs, toolCalling, visionInput

Benchmarks

  • Artificial Analysis: 9.9
  • OpenRouter Benchmarks: 48%

Qwen: Qwen3 VL 30B A3B Instruct

qwen/qwen3-vl-30b-a3b-instruct

Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Instruct variant optimizes instruction-following for general multimodal tasks. It excels in perception of real-world/synthetic categories, 2D/3

Context window
262,144
Input modalities
text, image
Output modalities
text
Capabilities
multimodal, structuredOutputs, toolCalling, visionInput

Benchmarks

  • Artificial Analysis: 7.9
  • OpenRouter Benchmarks: 38.667%

Inference prices

  • Input tokens (per 1M) : $0.29 / input_tokens_per_million · siliconFlow ↗
  • Output tokens (per 1M) : $1.00 / output_tokens_per_million · siliconFlow ↗

Qwen: Qwen3 VL 30B A3B Thinking

qwen/qwen3-vl-30b-a3b-thinking

Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos. Its Thinking variant enhances reasoning in STEM, math, and complex tasks. It excels...

Context window
262,144
Input modalities
text, image
Output modalities
text
Capabilities
multimodal, structuredOutputs, toolCalling, visionInput

Benchmarks

  • Artificial Analysis: 9.5

Inference prices

  • Input tokens (per 1M) : $0.29 / input_tokens_per_million · siliconFlow ↗
  • Output tokens (per 1M) : $1.00 / output_tokens_per_million · siliconFlow ↗

Qwen: Qwen3-VL-32B-Thinking

siliconFlow/Qwen/Qwen3-VL-32B-Thinking

Qwen3-VL-Thinking is a version of the Qwen3-VL series specially optimized for complex visual reasoning tasks. It incorporates a "Thinking Mode" , enabling it to generate detailed intermediate reasoning steps (Chain-of-Thought) before providing a final answer. This design signific

Maximum output
262,144
Input modalities
text, image
Output modalities
text
Capabilities
multimodal, structuredOutputs, toolCalling, visionInput

Inference prices

  • Input tokens (per 1M) : $0.20 / input_tokens_per_million · siliconFlow ↗
  • Output tokens (per 1M) : $1.50 / output_tokens_per_million · siliconFlow ↗

Qwen: Qwen3 VL 8B Instruct

qwen/qwen3-vl-8b-instruct

Qwen3-VL-8B-Instruct is a multimodal vision-language model from the Qwen3-VL series, built for high-fidelity understanding and reasoning across text, images, and video. It features improved multimodal fusion with Interleaved-MRoPE for long-horizon temporal reasoning, DeepStack fo

Context window
262,144
Input modalities
image, text
Output modalities
text
Capabilities
multimodal, promptCaching, structuredOutputs, toolCalling, visionInput

Benchmarks

  • Artificial Analysis: 7.3
  • OpenRouter Benchmarks: 32.667%

Inference prices

  • Input tokens (per 1M) : $0.18 / input_tokens_per_million · siliconFlow ↗
  • Output tokens (per 1M) : $0.68 / output_tokens_per_million · siliconFlow ↗

Qwen: Qwen-Image

siliconFlow/Qwen/Qwen-Image

Qwen-Image is an image generation foundation model released by the Alibaba Qwen team, featuring 20 billion parameters. The model has achieved significant advances in complex text rendering and precise image editing, excelling particularly at generating images with high-fidelity C

Input modalities
text
Output modalities
image
Capabilities
imageOutput

Benchmarks

  • LMArena: 1,056.989
  • MTEB: 98.436%

Inference prices

Qwen: Qwen-Image-Edit

siliconFlow/Qwen/Qwen-Image-Edit

Qwen-Image-Edit is the image editing version of Qwen-Image, released by Alibaba's Qwen team. Built upon the 20B Qwen-Image model, it has been further trained to extend its unique text rendering capabilities to image editing tasks, enabling precise text editing within images. Furt

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

Benchmarks

  • LMArena: 1,240.904

Inference prices

Qwen: Wan2.1-I2V-14B-720P

siliconFlow/Wan-AI/Wan2.1-I2V-14B-720P

Wan2.1-I2V-14B-720P is an open-source advanced image-to-video generation model, part of the Wan2.1 video foundation model suite. This 14B model can generate 720P high-definition videos. And after thousands of rounds of human evaluation, this model is reaching state-of-the-art per

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

Qwen: Wan2.1-I2V-14B-720P (Turbo)

siliconFlow/Wan-AI/Wan2.1-I2V-14B-720P-Turbo

Wan2.1-I2V-14B-720P-Turbo is the TeaCache accelerated version of the Wan2.1-I2V-14B-720P model, reducing single video generation time by 30%. Wan2.1-I2V-14B-720P is an open-source advanced image-to-video generation model, part of the Wan2.1 video foundation model suite. This 14B

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

Qwen: Wan2.1-T2V-14B

siliconFlow/Wan-AI/Wan2.1-T2V-14B

Wan2.1-T2V-14B is an open-source advanced text-to-video generation model. This 14B model has established state-of-the-art performance benchmarks among both open-source and closed-source models, capable of generating high-quality visual content with significant dynamic effects. It

Input modalities
text
Output modalities
video
Capabilities
videoOutput

Qwen: Wan2.1-T2V-14B (Turbo)

siliconFlow/Wan-AI/Wan2.1-T2V-14B-Turbo

Wan2.1-T2V-14B-T is the TeaCache accelerated version of the Wan2.1-T2V-14B model, reducing single video generation time by 30%. The Wan2.1-T2V-14B model has established state-of-the-art performance benchmarks among both open-source and closed-source models, capable of generating

Input modalities
text
Output modalities
video
Capabilities
videoOutput

Qwen: Wan2.2-I2V-A14B

siliconFlow/Wan-AI/Wan2.2-I2V-A14B

Wan2.2-I2V-A14B model is designed for image-to-video generation, supporting both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it achieves more stable video synthesis with reduced unrealistic camera movements and offers enhanced support for divers

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

Inference prices

Qwen: Wan2.2-T2V-A14B

siliconFlow/Wan-AI/Wan2.2-T2V-A14B

Wan2.2-T2V-A14B model supports generating 5s videos at both 480P and 720P resolutions. Built with a Mixture-of-Experts (MoE) architecture, it delivers outstanding video generation quality. On the new benchmark Wan-Bench 2.0, the model surpasses leading commercial models across mo

Input modalities
text
Output modalities
video
Capabilities
videoOutput

Inference prices

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