SiliconFlow

Inference host · 38 models · 38 priced routes · siliconflow.com

Models on SiliconFlow

About SiliconFlow

Across GlyphCompany profile Funding rounds socials

Model	In / out per 1M	Vs headline
Qwen3 Embedding 8B fp8 · The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks.	$0.04 / Free+300% vs headline	+300%	fp8 · The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks.
gpt-oss-20b fp8 · gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.	$0.04 / $0.18+934% vs headline	+934%	fp8 · gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.
Qwen3.5-9B fp8 · Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture.	$0.1 / $0.15+22% vs headline	+22%	fp8 · Qwen3.5-9B is a multimodal foundation model from the Qwen3.5 family, designed to deliver strong reasoning, coding, and visual understanding in an efficient 9B-parameter architecture.
Qwen3 Coder 30B A3B Instruct fp8 · Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.	$0.07 / $0.28Same vs headline	Same	fp8 · Qwen3-Coder-30B-A3B-Instruct is a 30.5B parameter Mixture-of-Experts (MoE) model with 128 experts (8 active per forward pass), designed for advanced code generation, repository-scale understanding, and agentic tool use.
Qwen3 30B A3B Instruct 2507 fp8 · Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.	$0.09 / $0.3Same vs headline	Same	fp8 · Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.
DeepSeek V4 Flash 0423 fp8 · DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.	$0.13 / $0.28+49% vs headline	+49%	fp8 · DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window.
Gemma 4 26B A4B fp8 · Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.	$0.14 / $0.4+141% vs headline	+141%	fp8 · Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.
GLM 5.3 Flash fp8 · GLM-5.3-Flash is a native multimodal model from Z.ai.	$0.15 / $0.5+389% vs headline	+389%	fp8 · GLM-5.3-Flash is a native multimodal model from Z.ai.
Qwen3 32B fp8 · Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.	$0.14 / $0.57+90% vs headline	+90%	fp8 · Qwen3-32B is a dense 32.8B parameter causal language model from the Qwen3 series, optimized for both complex reasoning and efficient dialogue.
gpt-oss-120b fp8 · gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.	$0.15 / $0.6+1467% vs headline	+1467%	fp8 · gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.
DeepSeek V3.2 Exp fp8 · DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.	$0.27 / $0.41Same vs headline	Same	fp8 · DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures.
GLM 4.5 Air fp8 · GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.	$0.14 / $0.86+3% vs headline	+3%	fp8 · GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications.
DeepSeek V4 Flash 0731 fp8 · DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.	$0.22 / $0.66+270% vs headline	+270%	fp8 · DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.
DeepSeek V3.2 fp8 · DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.	$0.259 / $0.8+36% vs headline	+36%	fp8 · DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance.
DeepSeek V3 0324 fp8 · DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.	$0.25 / $1Same vs headline	Same	fp8 · DeepSeek V3, a 685B-parameter, mixture-of-experts model, is the latest iteration of the flagship chat model family from the DeepSeek team.
DeepSeek V3.1 Terminus fp8 · DeepSeek-V3.1 Terminus is an update to DeepSeek V3.1 that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further	$0.27 / $1Same vs headline	Same	fp8 · DeepSeek-V3.1 Terminus is an update to DeepSeek V3.1 that maintains the model's original capabilities while addressing issues reported by users, including language consistency and agent capabilities, further
DeepSeek V3.1 fp8 · DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.	$0.27 / $1+6% vs headline	+6%	fp8 · DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.
Qwen3 VL 30B A3B Thinking fp8 · Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.	$0.29 / $1Same vs headline	Same	fp8 · Qwen3-VL-30B-A3B-Thinking is a multimodal model that unifies strong text generation with visual understanding for images and videos.
Qwen3 VL 30B A3B Instruct fp8 · Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.	$0.29 / $1+78% vs headline	+78%	fp8 · Qwen3-VL-30B-A3B-Instruct is a multimodal model that unifies strong text generation with visual understanding for images and videos.
DeepSeek V4.1 Flash fp8 · DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.	$0.3 / $1.2+383% vs headline	+383%	fp8 · DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.