API host · catalogue checked 27 Oct 2026

SiliconFlow: models, prices and uptime

SiliconFlow serves 13 of the 78 models we benchmark through OpenRouter. Prices are OpenRouter's list prices per million tokens. We did not measure speed per host.

Models served13of 78 we test

Cheapest host for2of 13 shared models

Median premium+167%over the cheapest host

Uptime92.6%last 30 min, average

Most accurate models on SiliconFlow

Verification score of each model this host serves

1DeepSeek V4.1 Flash 1017DeepSeek73.8%
2DeepSeek V4 Flash Vision ExpDeepSeek71.2%
3DeepSeek V4 Pro 0813DeepSeek71.2%
4GLM 5.3 FlashZ.ai67.5%
5GLM 5.3Z.ai67.5%
6Kimi K2.7 CodeMoonshotAI65.0%
7gpt-oss-120bOpenAI58.8%
8Qwen3.8 2.4T A95BQwen57.5%
9gpt-oss-20bOpenAI51.2%
10Qwen3.6 35B A3BQwen37.5%
11Gemma 4 31BGoogle30.0%
12Gemma 4 26B A4BGoogle15.0%

Models on SiliconFlow

Premium: this host's blended price (3 input : 1 output) against the cheapest host for the same model

Model	Score	Input / 1M	Output / 1M	vs cheapest host	Hosts	Context	Max output	Quantization	Uptime
DeepSeek V4.1 Flash 1017	73.8%	$0.3	$1.2	+364%	30	1M	393k	fp8	100%
DeepSeek V4 Flash Vision Exp	71.2%	$0.44	$1.32	+104%	3	1M	393k	fp8	100%
DeepSeek V4 Pro 0813	71.2%	$1.32	$3.96	+843%	23	1M	393k	fp8	100%
GLM 5.3 Flash	67.5%	$0.15	$0.5	+406%	34	1M	262k	fp8	93.3%
GLM 5.3	67.5%	$0.7	$2.2	+167%	37	1M	262k	fp8	100%
Kimi K2.7 Code	65.0%	$0.85916	$3.8	+33%	11	262k	236k	fp8	—
gpt-oss-120b	58.8%	$0.15	$0.6	+337%	23	131k	8k	fp8	45.5%
Qwen3.8 2.4T A95B	57.5%	$2	$6	Cheapest	5	1M	131k	fp8	100%
gpt-oss-20b	51.2%	$0.04	$0.18	+108%	10	131k	8k	fp8	100%
Qwen3.6 35B A3B	37.5%	$0.24	$1.8	+196%	10	262k	236k	fp8	100%
Gemma 4 31B	30.0%	$0.75	$1	+408%	13	262k	236k	fp8	72.7%
Gemma 4 26B A4B	15.0%	$0.14	$0.4	+137%	15	262k	236k	fp8	100%
Qwen3.5-122B-A10B	—	$0.26	$2.08	Cheapest	4	262k	236k	fp8	100%

All API hosts →

SiliconFlow: frequently asked questions

Which AI models does SiliconFlow serve?: Of the models we benchmark, SiliconFlow serves 13 through OpenRouter: DeepSeek V4.1 Flash 1017, DeepSeek V4 Flash Vision Exp, DeepSeek V4 Pro 0813, GLM 5.3 Flash, GLM 5.3, Kimi K2.7 Code, gpt-oss-120b, Qwen3.8 2.4T A95B, gpt-oss-20b, Qwen3.6 35B A3B, Gemma 4 31B, Gemma 4 26B A4B, Qwen3.5-122B-A10B.
What is the most accurate model on SiliconFlow?: DeepSeek V4.1 Flash 1017 by DeepSeek, with a verification score of 73.8% (59 of 80 questions correct) in our tests.
Is SiliconFlow the cheapest host?: SiliconFlow has the lowest list price for 2 of the 13 models it serves that other hosts also serve (Qwen3.8 2.4T A95B, Qwen3.5-122B-A10B). Its median markup over the cheapest host is +167%.
How reliable is SiliconFlow?: Averaged over the models it serves, SiliconFlow reported 92.6% uptime in the 30 minutes before OpenRouter's catalogue was read for our run.

Model

Score

Input / 1M

Output / 1M

vs cheapest host

Hosts

Context

Max output

Quantization

Uptime

DeepSeek V4.1 Flash 1017

73.8%

$0.3

$1.2

+364%

393k

fp8

100%

DeepSeek V4 Flash Vision Exp

71.2%

$0.44

$1.32

+104%

393k

fp8

100%

DeepSeek V4 Pro 0813

71.2%

$1.32

$3.96

+843%

393k

fp8

100%

GLM 5.3 Flash

67.5%

$0.15

$0.5

+406%

262k

fp8

93.3%

GLM 5.3

67.5%

$0.7

$2.2

+167%

262k

fp8

100%

Kimi K2.7 Code

65.0%

$0.85916

$3.8

+33%

262k

236k

fp8

—

gpt-oss-120b

58.8%

$0.15

$0.6

+337%

131k

fp8

45.5%

Qwen3.8 2.4T A95B

57.5%

Cheapest

131k

fp8

100%

gpt-oss-20b

51.2%

$0.04

$0.18

+108%

131k

fp8

100%

Qwen3.6 35B A3B

37.5%

$0.24

$1.8

+196%

262k

236k

fp8

100%

Gemma 4 31B

30.0%

$0.75

+408%

262k

236k

fp8

72.7%

Gemma 4 26B A4B

15.0%

$0.14

$0.4

+137%

262k

236k

fp8

100%

Qwen3.5-122B-A10B

—

$0.26

$2.08

Cheapest

262k

236k

fp8

100%

SiliconFlow: frequently asked questions

Which AI models does SiliconFlow serve?

Of the models we benchmark, SiliconFlow serves 13 through OpenRouter: DeepSeek V4.1 Flash 1017, DeepSeek V4 Flash Vision Exp, DeepSeek V4 Pro 0813, GLM 5.3 Flash, GLM 5.3, Kimi K2.7 Code, gpt-oss-120b, Qwen3.8 2.4T A95B, gpt-oss-20b, Qwen3.6 35B A3B, Gemma 4 31B, Gemma 4 26B A4B, Qwen3.5-122B-A10B.

What is the most accurate model on SiliconFlow?

DeepSeek V4.1 Flash 1017 by DeepSeek, with a verification score of 73.8% (59 of 80 questions correct) in our tests.

Is SiliconFlow the cheapest host?

SiliconFlow has the lowest list price for 2 of the 13 models it serves that other hosts also serve (Qwen3.8 2.4T A95B, Qwen3.5-122B-A10B). Its median markup over the cheapest host is +167%.

How reliable is SiliconFlow?

Averaged over the models it serves, SiliconFlow reported 92.6% uptime in the 30 minutes before OpenRouter's catalogue was read for our run.