Inference provider

SiliconFlow: AI models and API pricing

Every tracked model SiliconFlow serves through OpenRouter, with SiliconFlow's own input and output price per 1M tokens, the context window it serves, disclosed quantization and measured endpoint uptime. Rows link to the model page, where every provider's deployment is listed side by side.

Snapshot as of 2 October 2026

SiliconFlow serves 39 tracked models through 39 endpoints as of 2 October 2026; the highest-ranked is GLM 5.3 (ModelCap Index #11); mean measured uptime is 97.38%.

Models served

On the ModelCap Index

Cheapest output price

$0.15/1M

Language models on SiliconFlow (39)

Ranked models first, in ModelCap Index order. Prices are the lowest SiliconFlow lists for the model; the endpoint count shows how many deployments carry it.

Language models served by SiliconFlow, with price, context, quantization and uptime
#	Model	ModelCap Index	Input $/1M	Output $/1M	Context	Quantization	Uptime (30 min)	Endpoints
1	GLM 5.3Z.ai	#1180.9	$0.70	$2.20	1M	fp8	100.0%	1
2	DeepSeek V4.1 FlashDeepSeek	#1579.3	$0.30	$1.20	1M	fp8	100.0%	1Open weights
3	GLM 5.3 FlashZ.ai	#1778.9	$0.15	$0.50	1M	fp8	98.92%	1Open weights
4	DeepSeek V4 Pro 0813DeepSeek	#2476.4	$1.32	$3.96	1M	fp8	100.0%	1Open weights
5	Hy4 previewTencent	#2673.8	$0.834	$2.50	1M	fp8	98.17%	1Open weights
6	DeepSeek V4 Flash Vision ExpDeepSeek	#2773.8	$0.44	$1.32	1M	fp8	99.72%	1Open weights
7	Gemma 4 31BGoogle	#2872.9	$0.75	$1.00	262K	fp8	100.0%	1Open weights
8	Gemma 4 26B A4BGoogle	#4168.4	$0.14	$0.40	262K	fp8	97.69%	1Open weights
9	Kimi K2.7 CodeMoonshot AI	#4965.9	$0.859	$3.80	262K	fp8	100.0%	1
10	DeepSeek V3.2DeepSeek	#6661.7	$0.259	$0.42	164K	fp8	81.01%	1Open weights
11	R1 0528DeepSeek	#6761.5	$0.50	$2.18	164K	fp8	99.09%	1Open weights
12	Qwen3.6 35B A3BQwen	#7260.7	$0.24	$1.80	262K	fp8	98.51%	1Open weights
13	DeepSeek V3.1 TerminusDeepSeek	#7659.7	$0.27	$1.00	164K	fp8	99.47%	1Open weights
14	Qwen3.5-122B-A10BQwen	#7759.5	$0.26	$2.08	262K	fp8	100.0%	1Open weights
15	Qwen3.5-9BQwen	#9554.9	$0.10	$0.15	262K	fp8	99.78%	1Open weights
16	Qwen3 Coder 30B A3B InstructQwen	#9954.1	$0.07	$0.28	262K	fp8	100.0%	1Open weights
17	Qwen3 30B A3B Instruct 2507Qwen	#12050.6	$0.09	$0.30	262K	fp8	99.29%	1Open weights
18	GLM 4.5 AirZ.ai	#12547.4	$0.14	$0.86	131K	fp8	100.0%	1Open weights
19	gpt-oss-120bOpenAI	#13244.6	$0.15	$0.60	131K	fp8	65.41%	1Open weights
20	Qwen3 32BQwen	#14441.2	$0.14	$0.57	131K	fp8	98.04%	1Open weights
21	gpt-oss-20bOpenAI	#16637.8	$0.04	$0.18	131K	fp8	94.17%	1Open weights
22	Hunyuan A13B InstructTencent	—	$0.14	$0.57	131K	fp8	100.0%	1
23	Qwen3 VL 30B A3B InstructQwen	—	$0.29	$1.00	262K	fp8	77.95%	1Open weights
24	Qwen3 VL 30B A3B ThinkingQwen	—	$0.29	$1.00	262K	fp8	—	1Open weights
25	DeepSeek V3 0324DeepSeek	—	$0.25	$1.00	164K	fp8	99.59%	1Open weights
26	DeepSeek V3.1DeepSeek	—	$0.27	$1.00	164K	fp8	99.76%	1Open weights
27	DeepSeek V3.2 ExpDeepSeek	—	$0.27	$0.41	164K	fp8	99.81%	1Open weights
28	DeepSeek V4 Flash 0423DeepSeek	—	$0.13	$0.28	1M	fp8	99.94%	1Open weights
29	DeepSeek V4 Flash 0731DeepSeek	—	$0.22	$0.66	1M	fp8	99.82%	1Open weights
30	DeepSeek V4 Pro 0423DeepSeek	—	$1.50	$3.14	1M	fp8	99.83%	1Open weights
31	GLM 5Z.ai	—	$0.95	$2.55	205K	fp8	99.65%	1Open weights
32	GLM 5.1Z.ai	—	$1.19	$3.74	205K	fp8	98.60%	1Open weights
33	GLM 5.2Z.ai	—	$0.70	$2.20	1M	fp8	100.0%	1Open weights
34	Kimi K2.5Moonshot AI	—	$0.45	$2.25	262K	int4	99.85%	1
35	Kimi K2.6Moonshot AI	—	$0.77	$3.40	262K	fp8	100.0%	1
36	Qwen3.5-27BQwen	—	$0.25	$2.00	262K	fp8	100.0%	1Open weights
37	Qwen3.5-35B-A3BQwen	—	$0.24	$1.80	262K	fp8	—	1Open weights
38	Qwen3.6 27BQwen	—	$0.30	$3.20	262K	fp8	99.08%	1Open weights
39	Step 3.5 FlashStepFun	—	$0.10	$0.30	262K	fp8	100.0%	1Open weights

SiliconFlow on ModelCap: common questions

How many AI models does SiliconFlow serve?

39 tracked models as of 2 October 2026: 39 language models (21 with a public ModelCap Index position), across 39 live endpoints observed through OpenRouter.

What is the best model available on SiliconFlow?

GLM 5.3 from Z.ai is the highest-ranked language model SiliconFlow serves, at ModelCap Index position #11 and $2.20 per 1M output tokens on SiliconFlow. The table lists every model in Index order.

What is the cheapest model on SiliconFlow?

Qwen3.5-9B at $0.15 per 1M output tokens ($0.10 per 1M input), the lowest positive language-token price SiliconFlow lists. Free-tier endpoints, where present, show as $0 and are excluded from that comparison.

Which SiliconFlow model has the largest context window?

GLM 5.3, served with a 1M-token context window on SiliconFlow. Providers sometimes serve less than the model's published maximum, so the figure here is the provider's own.

How reliable is SiliconFlow?

Mean measured uptime across SiliconFlow endpoints is 97.38% over the last 30 minutes, as reported by OpenRouter at the time of the snapshot. Per-model uptime is in the table; it is a live operational signal, not a service-level guarantee.

Where do the SiliconFlow prices come from?

From the OpenRouter catalogue entry for each SiliconFlow endpoint, in USD per 1M tokens, re-observed on every ModelCap refresh. Where SiliconFlow exposes several deployments of one model at different prices, the table shows the lowest and notes the count; the model page lists each deployment.

How current is this page?

This page uses the published ModelCap snapshot dated 2 October 2026. Publication and page updates happen separately. Offers retained from a failed refresh are marked stale in the data and are not counted here.