Provides cost-effective generative AI cloud services based on open-source models for text, image, video, and audio generation.

Qwen3 Qwen3-VL Qwen3-Next Qwen2 GLM-4 GLM-4.5 GLM-4.5V Qwen2.5 Qwen2.5-Coder DeepSeek-V3.1 DeepSeek-V3 Qwen2.5-VL DeepSeek-V3.2 DeepSeek-V2 DeepSeek-V2.5

SiliconFlow offers generative AI cloud services built on open-source models. Primary capabilities include text conversation, image generation, video generation, and speech synthesis. The platform emphasizes high cost-effectiveness for these AI generation tasks. Typical use cases involve leveraging these multimodal AI services for various creative and conversational applications.

Model	Speed	Latency	Tests
THUDM/GLM-Z1-9B-0414	171.50 t/s	13.03s	25
deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B	142.95 t/s	4.47s	5
Qwen/Qwen3-VL-8B-Instruct	142.73 t/s	0.64s	5
Qwen/Qwen3-VL-8B-Instruct	142.73 t/s	0.64s	5
tencent/Hunyuan-MT-7B	88.25 t/s	0.51s	5
Qwen/Qwen3-Next-80B-A3B-Instruct	86.82 t/s	0.62s	10
Qwen/Qwen3-Next-80B-A3B-Instruct	86.82 t/s	0.62s	10
Qwen/Qwen2-7B-Instruct	81.93 t/s	0.57s	25
Qwen/Qwen3-14B	78.64 t/s	9.81s	5
Pro/THUDM/glm-4-9b-chat	76.25 t/s	0.63s	10
THUDM/glm-4-9b-chat	75.38 t/s	0.59s	15
zai-org/GLM-4.5V	73.00 t/s	6.15s	10
zai-org/GLM-4.5V	73.00 t/s	6.15s	10
zai-org/GLM-4.5V	73.00 t/s	6.15s	10
Pro/Qwen/Qwen2-7B-Instruct	71.29 t/s	0.56s	5
Qwen/QwQ-32B-Preview	69.75 t/s	0.62s	5
internlm/internlm2_5-7b-chat	68.17 t/s	0.56s	20
Qwen/Qwen2.5-Coder-7B-Instruct	68.13 t/s	0.47s	30
Qwen/Qwen2.5-Coder-7B-Instruct	68.13 t/s	0.47s	30
Qwen/Qwen2.5-Coder-7B-Instruct	68.13 t/s	0.47s	30
deepseek-ai/DeepSeek-R1-Distill-Qwen-14B	67.68 t/s	10.66s	5
Pro/THUDM/GLM-4.1V-9B-Thinking	64.99 t/s	15.30s	5
LoRA/Qwen/Qwen2.5-32B-Instruct	64.68 t/s	0.84s	5
LoRA/Qwen/Qwen2.5-32B-Instruct	64.68 t/s	0.84s	5
Qwen/Qwen2.5-14B-Instruct	64.55 t/s	0.61s	5
Qwen/Qwen2.5-14B-Instruct	64.55 t/s	0.61s	5
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B	61.74 t/s	12.60s	20
LoRA/Qwen/Qwen2.5-14B-Instruct	58.12 t/s	1.02s	5
LoRA/Qwen/Qwen2.5-14B-Instruct	58.12 t/s	1.02s	5
Qwen/Qwen2.5-32B-Instruct	57.83 t/s	0.69s	15
Qwen/Qwen2.5-32B-Instruct	57.83 t/s	0.69s	15
deepseek-ai/DeepSeek-R1-0528-Qwen3-8B	53.62 t/s	28.33s	30
THUDM/GLM-4.1V-9B-Thinking	46.12 t/s	23.86s	20
Qwen/Qwen3-30B-A3B	41.78 t/s	37.81s	15
zai-org/GLM-4.5-Air	36.90 t/s	2.05s	10
zai-org/GLM-4.5-Air	36.90 t/s	2.05s	10
deepseek-ai/DeepSeek-V3.1-Terminus	36.65 t/s	2.17s	20
deepseek-ai/DeepSeek-V3.1-Terminus	36.65 t/s	2.17s	20
Qwen/Qwen3-8B	36.27 t/s	41.32s	55
Qwen/QwQ-32B	35.94 t/s	35.43s	30
deepseek-ai/DeepSeek-V3.1	35.75 t/s	3.81s	10
deepseek-ai/DeepSeek-V3.1	35.75 t/s	3.81s	10
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	31.03 t/s	40.29s	10
Qwen/Qwen2.5-7B-Instruct	29.82 t/s	1.68s	25
Qwen/Qwen2.5-7B-Instruct	29.82 t/s	1.68s	25
Qwen/Qwen2.5-72B-Instruct	28.66 t/s	1.04s	30
Qwen/Qwen2.5-72B-Instruct	28.66 t/s	1.04s	30
Pro/Qwen/Qwen2.5-Coder-7B-Instruct	28.32 t/s	0.63s	5
Pro/Qwen/Qwen2.5-Coder-7B-Instruct	28.32 t/s	0.63s	5
Pro/Qwen/Qwen2.5-Coder-7B-Instruct	28.32 t/s	0.63s	5
Vendor-A/Qwen/Qwen2.5-72B-Instruct	27.74 t/s	1.25s	5
Vendor-A/Qwen/Qwen2.5-72B-Instruct	27.74 t/s	1.25s	5
Qwen/Qwen2.5-VL-72B-Instruct	26.42 t/s	0.86s	5
Qwen/Qwen2.5-VL-72B-Instruct	26.42 t/s	0.86s	5
Qwen/Qwen2.5-VL-72B-Instruct	26.42 t/s	0.86s	5
LoRA/Qwen/Qwen2.5-72B-Instruct	24.64 t/s	3.37s	5
LoRA/Qwen/Qwen2.5-72B-Instruct	24.64 t/s	3.37s	5
Pro/deepseek-ai/DeepSeek-V3.2-Exp	24.18 t/s	0.98s	5
Pro/deepseek-ai/DeepSeek-V3.2-Exp	24.18 t/s	0.98s	5
deepseek-ai/DeepSeek-V3.2-Exp	23.77 t/s	2.51s	25
deepseek-ai/DeepSeek-V3.2-Exp	23.77 t/s	2.51s	25
Qwen/Qwen3-235B-A22B	23.66 t/s	40.92s	15
Qwen/Qwen2.5-Coder-32B-Instruct	23.58 t/s	1.18s	5
Qwen/Qwen2.5-Coder-32B-Instruct	23.58 t/s	1.18s	5
Qwen/Qwen2.5-Coder-32B-Instruct	23.58 t/s	1.18s	5
deepseek-ai/DeepSeek-R1-Distill-Llama-70B	22.73 t/s	35.98s	10
deepseek-ai/DeepSeek-V3	22.00 t/s	1.92s	165
Qwen/Qwen2.5-72B-Instruct-128K	18.27 t/s	0.86s	5
Qwen/Qwen2.5-72B-Instruct-128K	18.27 t/s	0.86s	5
moonshotai/Kimi-K2-Instruct	17.07 t/s	1.44s	15
Qwen/Qwen3-235B-A22B-Instruct-2507	15.82 t/s	0.75s	5
deepseek-ai/DeepSeek-R1	15.45 t/s	55.46s	50
deepseek-ai/DeepSeek-V2.5	15.00 t/s	0.79s	5
deepseek-ai/DeepSeek-V2.5	15.00 t/s	0.79s	5
inclusionAI/Ling-1T	6.44 t/s	0.93s	5
Pro/deepseek-ai/DeepSeek-R1	0.00 t/s	0.00s	5
Qwen3-235B-A22B	0.00 t/s	0.00s	5
qwen2.5	0.00 t/s	0.00s	5
qwen2.5	0.00 t/s	0.00s	5

Time	Model	Speed	Latency
Dec 8, 04:34 PM	Unknown	-	-s
Dec 4, 02:14 PM	deepseek-ai/DeepSeek-V3.1-Terminus	18.32 t/s	2.18s
Nov 3, 07:41 PM	Qwen/Qwen3-8B	42.83 t/s	16.85s
Nov 1, 02:00 AM	Unknown	-	-s
Nov 1, 01:55 AM	Qwen/Qwen2.5-Coder-7B-Instruct	97.24 t/s	0.29s
Nov 1, 01:53 AM	Qwen/Qwen2.5-Coder-7B-Instruct	96.70 t/s	0.23s
Nov 1, 01:33 AM	Qwen/Qwen2.5-Coder-7B-Instruct	93.70 t/s	0.23s
Oct 25, 02:13 PM	deepseek-ai/DeepSeek-V3.2-Exp	22.98 t/s	1.27s
Oct 25, 01:14 PM	deepseek-ai/DeepSeek-V3.2-Exp	23.66 t/s	1.65s
Oct 25, 12:53 PM	zai-org/GLM-4.5V	72.53 t/s	6.41s

Model	Speed	Latency	Tests
THUDM/GLM-Z1-9B-0414	171.50 t/s	13.03s	25
deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B	142.95 t/s	4.47s	5
Qwen/Qwen3-VL-8B-Instruct	142.73 t/s	0.64s	5
Qwen/Qwen3-VL-8B-Instruct	142.73 t/s	0.64s	5
tencent/Hunyuan-MT-7B	88.25 t/s	0.51s	5
Qwen/Qwen3-Next-80B-A3B-Instruct	86.82 t/s	0.62s	10
Qwen/Qwen3-Next-80B-A3B-Instruct	86.82 t/s	0.62s	10
Qwen/Qwen2-7B-Instruct	81.93 t/s	0.57s	25
Qwen/Qwen3-14B	78.64 t/s	9.81s	5
Pro/THUDM/glm-4-9b-chat	76.25 t/s	0.63s	10
THUDM/glm-4-9b-chat	75.38 t/s	0.59s	15
zai-org/GLM-4.5V	73.00 t/s	6.15s	10
zai-org/GLM-4.5V	73.00 t/s	6.15s	10
zai-org/GLM-4.5V	73.00 t/s	6.15s	10
Pro/Qwen/Qwen2-7B-Instruct	71.29 t/s	0.56s	5
Qwen/QwQ-32B-Preview	69.75 t/s	0.62s	5
internlm/internlm2_5-7b-chat	68.17 t/s	0.56s	20
Qwen/Qwen2.5-Coder-7B-Instruct	68.13 t/s	0.47s	30
Qwen/Qwen2.5-Coder-7B-Instruct	68.13 t/s	0.47s	30
Qwen/Qwen2.5-Coder-7B-Instruct	68.13 t/s	0.47s	30
deepseek-ai/DeepSeek-R1-Distill-Qwen-14B	67.68 t/s	10.66s	5
Pro/THUDM/GLM-4.1V-9B-Thinking	64.99 t/s	15.30s	5
LoRA/Qwen/Qwen2.5-32B-Instruct	64.68 t/s	0.84s	5
LoRA/Qwen/Qwen2.5-32B-Instruct	64.68 t/s	0.84s	5
Qwen/Qwen2.5-14B-Instruct	64.55 t/s	0.61s	5
Qwen/Qwen2.5-14B-Instruct	64.55 t/s	0.61s	5
deepseek-ai/DeepSeek-R1-Distill-Qwen-7B	61.74 t/s	12.60s	20
LoRA/Qwen/Qwen2.5-14B-Instruct	58.12 t/s	1.02s	5
LoRA/Qwen/Qwen2.5-14B-Instruct	58.12 t/s	1.02s	5
Qwen/Qwen2.5-32B-Instruct	57.83 t/s	0.69s	15
Qwen/Qwen2.5-32B-Instruct	57.83 t/s	0.69s	15
deepseek-ai/DeepSeek-R1-0528-Qwen3-8B	53.62 t/s	28.33s	30
THUDM/GLM-4.1V-9B-Thinking	46.12 t/s	23.86s	20
Qwen/Qwen3-30B-A3B	41.78 t/s	37.81s	15
zai-org/GLM-4.5-Air	36.90 t/s	2.05s	10
zai-org/GLM-4.5-Air	36.90 t/s	2.05s	10
deepseek-ai/DeepSeek-V3.1-Terminus	36.65 t/s	2.17s	20
deepseek-ai/DeepSeek-V3.1-Terminus	36.65 t/s	2.17s	20
Qwen/Qwen3-8B	36.27 t/s	41.32s	55
Qwen/QwQ-32B	35.94 t/s	35.43s	30
deepseek-ai/DeepSeek-V3.1	35.75 t/s	3.81s	10
deepseek-ai/DeepSeek-V3.1	35.75 t/s	3.81s	10
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	31.03 t/s	40.29s	10
Qwen/Qwen2.5-7B-Instruct	29.82 t/s	1.68s	25
Qwen/Qwen2.5-7B-Instruct	29.82 t/s	1.68s	25
Qwen/Qwen2.5-72B-Instruct	28.66 t/s	1.04s	30
Qwen/Qwen2.5-72B-Instruct	28.66 t/s	1.04s	30
Pro/Qwen/Qwen2.5-Coder-7B-Instruct	28.32 t/s	0.63s	5
Pro/Qwen/Qwen2.5-Coder-7B-Instruct	28.32 t/s	0.63s	5
Pro/Qwen/Qwen2.5-Coder-7B-Instruct	28.32 t/s	0.63s	5
Vendor-A/Qwen/Qwen2.5-72B-Instruct	27.74 t/s	1.25s	5
Vendor-A/Qwen/Qwen2.5-72B-Instruct	27.74 t/s	1.25s	5
Qwen/Qwen2.5-VL-72B-Instruct	26.42 t/s	0.86s	5
Qwen/Qwen2.5-VL-72B-Instruct	26.42 t/s	0.86s	5
Qwen/Qwen2.5-VL-72B-Instruct	26.42 t/s	0.86s	5
LoRA/Qwen/Qwen2.5-72B-Instruct	24.64 t/s	3.37s	5
LoRA/Qwen/Qwen2.5-72B-Instruct	24.64 t/s	3.37s	5
Pro/deepseek-ai/DeepSeek-V3.2-Exp	24.18 t/s	0.98s	5
Pro/deepseek-ai/DeepSeek-V3.2-Exp	24.18 t/s	0.98s	5
deepseek-ai/DeepSeek-V3.2-Exp	23.77 t/s	2.51s	25
deepseek-ai/DeepSeek-V3.2-Exp	23.77 t/s	2.51s	25
Qwen/Qwen3-235B-A22B	23.66 t/s	40.92s	15
Qwen/Qwen2.5-Coder-32B-Instruct	23.58 t/s	1.18s	5
Qwen/Qwen2.5-Coder-32B-Instruct	23.58 t/s	1.18s	5
Qwen/Qwen2.5-Coder-32B-Instruct	23.58 t/s	1.18s	5
deepseek-ai/DeepSeek-R1-Distill-Llama-70B	22.73 t/s	35.98s	10
deepseek-ai/DeepSeek-V3	22.00 t/s	1.92s	165
Qwen/Qwen2.5-72B-Instruct-128K	18.27 t/s	0.86s	5
Qwen/Qwen2.5-72B-Instruct-128K	18.27 t/s	0.86s	5
moonshotai/Kimi-K2-Instruct	17.07 t/s	1.44s	15
Qwen/Qwen3-235B-A22B-Instruct-2507	15.82 t/s	0.75s	5
deepseek-ai/DeepSeek-R1	15.45 t/s	55.46s	50
deepseek-ai/DeepSeek-V2.5	15.00 t/s	0.79s	5
deepseek-ai/DeepSeek-V2.5	15.00 t/s	0.79s	5
inclusionAI/Ling-1T	6.44 t/s	0.93s	5
Pro/deepseek-ai/DeepSeek-R1	0.00 t/s	0.00s	5
Qwen3-235B-A22B	0.00 t/s	0.00s	5
qwen2.5	0.00 t/s	0.00s	5
qwen2.5	0.00 t/s	0.00s	5

Time	Model	Speed	Latency
Dec 8, 04:34 PM	Unknown	-	-s
Dec 4, 02:14 PM	deepseek-ai/DeepSeek-V3.1-Terminus	18.32 t/s	2.18s
Nov 3, 07:41 PM	Qwen/Qwen3-8B	42.83 t/s	16.85s
Nov 1, 02:00 AM	Unknown	-	-s
Nov 1, 01:55 AM	Qwen/Qwen2.5-Coder-7B-Instruct	97.24 t/s	0.29s
Nov 1, 01:53 AM	Qwen/Qwen2.5-Coder-7B-Instruct	96.70 t/s	0.23s
Nov 1, 01:33 AM	Qwen/Qwen2.5-Coder-7B-Instruct	93.70 t/s	0.23s
Oct 25, 02:13 PM	deepseek-ai/DeepSeek-V3.2-Exp	22.98 t/s	1.27s
Oct 25, 01:14 PM	deepseek-ai/DeepSeek-V3.2-Exp	23.66 t/s	1.65s
Oct 25, 12:53 PM	zai-org/GLM-4.5V	72.53 t/s	6.41s

SiliconFlow

Supported Models

Recent Test Records

SiliconFlow

Supported Models

Recent Test Records

SiliconFlow

About SiliconFlow

Supported Models

Recent Test Records

SiliconFlow

About SiliconFlow

Supported Models

Recent Test Records