SiliconFlow API pricing
Updated
SiliconFlow lists standard per-token prices for 56 models. The cheapest is Qwen 2.5 7B at USD 0.05 input and USD 0.05 output per 1M tokens; the dearest is Kimi K3 at USD 2.70 / USD 13.50. It publishes cached-input prices for 26 of them, and lists no per-model batch, flex or priority prices on the pages we read. Checked 2026-11-15.
- Models with a standard price
- 56
- Median input / output
- USD 0.25 / USD 1
- Cached-input prices
- 26 of 56
- Cheapest host for
- 11 of 47 shared models
Which models is SiliconFlow the cheapest provider for?
Of the 47 models that SiliconFlow and at least one other provider both price at the standard tier, SiliconFlow has the lowest price for 11, ranked on a blend of 3 input tokens to 1 output token. It also ties the lowest price for 6 more, where another provider is listed first. It is the dearest for 4. Another 9 models have a standard price from SiliconFlow only.
| Model | SiliconFlow input / output | Next cheapest | Saving on 3:1 blend | Providers |
|---|---|---|---|---|
| Kimi K2.5 | USD 0.45 / USD 2.25 | Amazon Bedrock USD 0.60 / USD 3 | 25% | 8 |
| Qwen3 Coder 480B A35B | USD 0.25 / USD 1 | DeepInfra USD 0.30 / USD 1 | 8% | 6 |
| Qwen3.5 122B A10B | USD 0.26 / USD 2.08 | Alibaba Cloud USD 0.40 / USD 3.20 | 35% | 4 |
| Hy3 | USD 0.132 / USD 0.528 | GMI Cloud USD 0.14 / USD 0.58 | 8% | 4 |
| Qwen3 Coder 30B A3B | USD 0.07 / USD 0.28 | Featherless USD 0.125 / USD 0.45 | 41% | 3 |
| Qwen 2.5 7B | USD 0.05 / USD 0.05 | Venice USD 0.05 / USD 0.13 | 29% | 3 |
| Qwen3-14B | USD 0.07 / USD 0.28 | DeepInfra USD 0.12 / USD 0.24 | 18% | 3 |
| DeepSeek V3 | USD 0.25 / USD 1 | DeepInfra USD 0.32 / USD 0.89 | 5% | 3 |
| Qwen3-8B | USD 0.06 / USD 0.06 | Featherless USD 0.117 / USD 0.455 | 70% | 2 |
| Qwen3-30B-A3B-Instruct-2507 | USD 0.09 / USD 0.30 | Featherless USD 0.125 / USD 0.45 | 31% | 2 |
| Qwen3-VL 8B | USD 0.18 / USD 0.68 | Parasail USD 0.25 / USD 0.75 | 19% | 2 |
How does SiliconFlow price the most widely hosted models?
The 6 models SiliconFlow sells that the most other providers also sell, with SiliconFlow's rank among them on the 3:1 blend.
| Model | SiliconFlow | Rank | Cheapest |
|---|---|---|---|
| GLM 5.3 | USD 1.40 / USD 4.40 | 6 of 24 | DeepInfra USD 0.5625 / USD 2.50 |
| OpenAI GPT OSS 120B | USD 0.05 / USD 0.45 | 7 of 21 | CoreWeave Forge USD 0.03 / USD 0.17 |
| GLM 5.3 Flash | USD 0.15 / USD 0.50 | 3 of 20 | GMI Cloud USD 0.075 / USD 0.25 |
| GLM 5.2 | USD 1.302 / USD 4.092 | 4 of 19 | DeepInfra USD 0.5625 / USD 1.80 |
| Kimi K3 | USD 2.70 / USD 13.50 | 3 of 18 | Chutes USD 1.85 / USD 9.50 |
| DeepSeek V4.1 Flash | USD 0.15 / USD 0.60 | 3 of 16 | DeepInfra USD 0.14 / USD 0.42 |
Which model makers does SiliconFlow sell?
SiliconFlow prices models from 12 makers. Qwen leads with 20 models. Standard prices, USD per 1M input / output tokens.
| Maker | Models | Cheapest model | Median input / output |
|---|---|---|---|
| Qwen | 20 | Qwen 2.5 7B USD 0.05 / USD 0.05 | USD 0.22 / USD 0.84 |
| DeepSeek | 12 | DeepSeek V4 Flash USD 0.13 / USD 0.28 | USD 0.27 / USD 1 |
| Z.ai | 7 | GLM 5.3 Flash USD 0.15 / USD 0.50 | USD 1.19 / USD 3.74 |
| Moonshot AI | 4 | Kimi K2.5 USD 0.45 / USD 2.25 | USD 0.81458 / USD 3.60 |
| 3 | gemma-4-12B-it USD 0.10 / USD 0.30 | USD 0.14 / USD 0.40 | |
| Tencent | 3 | Hy3 USD 0.132 / USD 0.528 | USD 0.14 / USD 0.57 |
| OpenAI | 2 | GPT OSS 20B USD 0.04 / USD 0.18 | USD 0.045 / USD 0.315 |
| ByteDance Seed | 1 | Seed-OSS-36B-Instruct USD 0.21 / USD 0.57 | — |
| inclusionAI | 1 | Ling-flash-2.0 USD 0.14 / USD 0.57 | — |
| Meituan | 1 | LongCat-2.0 USD 0.75 / USD 2.95 | — |
| MiniMax | 1 | MiniMax M3 USD 0.30 / USD 1.20 | — |
| StepFun | 1 | Step-3.5-Flash USD 0.10 / USD 0.30 | — |
SiliconFlow price list
USD per 1M tokens: 56 prices for 56 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-15 unless a row says otherwise. "vs maker" compares a standard price with the model maker's own API on the 3:1 blend; a minus means SiliconFlow is cheaper. Of 19 hosted model prices with a maker price to compare, 14 are below the maker's and 0 above.
| Model | Tier | Input | Cached | Output | Context | vs maker | Source |
|---|---|---|---|---|---|---|---|
| DeepSeek R1 | standard | USD 0.50 | — | USD 2.18 | 163,840 | — | source |
| DeepSeek V3 | standard | USD 0.25 | — | USD 1 | 163,840 | — | source 2026-11-14 |
| DeepSeek V3.1 | standard | USD 0.27 | — | USD 1 | 163,840 | — | source 2026-11-14 |
| DeepSeek V3.2 | standard | USD 0.27 | USD 0.135 | USD 0.42 | 163,840 | — | source 2026-11-14 |
| DeepSeek V4 Flash | standard | USD 0.13 | USD 0.028 | USD 0.28 | 1,048,576 | — | source 2026-11-14 |
| DeepSeek V4 Flash 0731 | standard | USD 0.22 | USD 0.014 | USD 0.66 | 1,048,576 | — | source 2026-11-14 |
| DeepSeek V4 Pro | standard | USD 1.50162 | USD 0.135 | USD 3.135 | 1,048,576 | −4% | source 2026-11-14 |
| DeepSeek V4 Pro 0813 | standard | USD 1.32 | USD 0.044 | USD 3.96 | 1,048,576 | same | source 2026-11-14 |
| DeepSeek V4.1 Flash | standard | USD 0.15 | USD 0.003 | USD 0.60 | 1,048,576 | −50% | source 2026-11-14 |
| DeepSeek-V3.1-Terminus | standard | USD 0.27 | — | USD 1 | 163,840 | — | source 2026-11-14 |
| DeepSeek-V3.2-Exp | standard | USD 0.27 | — | USD 0.41 | 163,840 | — | source 2026-11-14 |
| DeepSeek-V4-Flash-Vision-Exp | standard | USD 0.44 | USD 0.028 | USD 1.32 | 1,048,576 | — | source 2026-11-14 |
| Gemma 4 26B A4B | standard | USD 0.14 | USD 0.05 | USD 0.40 | 262,144 | — | source 2026-11-14 |
| Gemma 4 31B | standard | USD 0.75 | USD 0.25 | USD 1 | 262,144 | — | source 2026-11-14 |
| gemma-4-12B-it | standard | USD 0.10 | — | USD 0.30 | 262,144 | — | source 2026-11-14 |
| GLM 4.5 Air | standard | USD 0.14 | — | USD 0.86 | 131,072 | −25% | source 2026-11-14 |
| GLM 5 | standard | USD 0.95 | USD 0.20 | USD 2.55 | 204,800 | −13% | source 2026-11-14 |
| GLM 5.1 | standard | USD 1.19 | USD 0.60 | USD 3.74 | 204,800 | −15% | source 2026-11-14 |
| GLM 5.2 | standard | USD 1.302 | USD 0.26 | USD 4.092 | 1,048,576 | −7% | source 2026-11-14 |
| GLM 5.3 | standard | USD 1.40 | USD 0.26 | USD 4.40 | 1,048,576 | same | source 2026-11-14 |
| GLM 5.3 Flash | standard | USD 0.15 | USD 0.03 | USD 0.50 | 1,048,576 | same | source 2026-11-14 |
| GLM-5V-Turbo | standard | USD 1.20 | USD 0.24 | USD 4 | 204,800 | — | source 2026-11-14 |
| GPT OSS 20B | standard | USD 0.04 | — | USD 0.18 | 131,072 | — | source 2026-11-14 |
| Hunyuan-A13B-Instruct | standard | USD 0.14 | — | USD 0.57 | 131,072 | — | source 2026-11-14 |
| Hy3 | standard | USD 0.132 | USD 0.033 | USD 0.528 | 262,144 | — | source 2026-11-14 |
| Hy4 Preview | standard | USD 0.834 | USD 0.042 | USD 2.501 | 1,048,576 | — | source |
| Kimi K2.5 | standard | USD 0.45 | USD 0.07 | USD 2.25 | 262,144 | — | source 2026-11-14 |
| Kimi K2.6 | standard | USD 0.77 | USD 0.14 | USD 3.40 | 262,144 | −17% | source 2026-11-14 |
| Kimi K2.7 Code | standard | USD 0.85916 | USD 0.17993 | USD 3.80 | 262,144 | −7% | source 2026-11-14 |
| Kimi K3 | standard | USD 2.70 | USD 0.27 | USD 13.50 | 1,048,576 | −10% | source 2026-11-14 |
| Ling-flash-2.0 | standard | USD 0.14 | — | USD 0.57 | 131,072 | — | source 2026-11-14 |
| LongCat-2.0 | standard | USD 0.75 | USD 0.015 | USD 2.95 | 1,048,576 | — | source 2026-11-14 |
| MiniMax M3 | standard | USD 0.30 | USD 0.06 | USD 1.20 | 1,048,576 | same | source 2026-11-14 |
| OpenAI GPT OSS 120B | standard | USD 0.05 | USD 0.075 | USD 0.45 | 131,072 | — | source 2026-11-14 |
| Qwen 2.5 7B | standard | USD 0.05 | — | USD 0.05 | 32,768 | — | source 2026-11-14 |
| Qwen2.5-72B-Instruct | standard | USD 0.59 | — | USD 0.59 | 32,768 | — | source 2026-11-14 |
| Qwen3 32B FP8 | standard | USD 0.14 | — | USD 0.57 | 131,072 | — | source 2026-11-14 |
| Qwen3 Coder 30B A3B | standard | USD 0.07 | — | USD 0.28 | 262,144 | — | source 2026-11-14 |
| Qwen3 Coder 480B A35B | standard | USD 0.25 | — | USD 1 | 262,144 | — | source 2026-11-14 |
| Qwen3 VL 30B A3B Instruct | standard | USD 0.29 | — | USD 1 | 262,144 | — | source 2026-11-14 |
| Qwen3-14B | standard | USD 0.07 | — | USD 0.28 | 131,072 | — | source 2026-11-14 |
| Qwen3-30B-A3B-Instruct-2507 | standard | USD 0.09 | — | USD 0.30 | 262,144 | — | source 2026-11-14 |
| Qwen3-8B | standard | USD 0.06 | — | USD 0.06 | 131,072 | — | source 2026-11-14 |
| Qwen3-VL 8B | standard | USD 0.18 | — | USD 0.68 | 262,144 | — | source 2026-11-14 |
| Qwen3-VL-30B-A3B-Thinking | standard | USD 0.29 | — | USD 1 | 262,144 | — | source 2026-11-14 |
| Qwen3-VL-32B-Instruct | standard | USD 0.20 | — | USD 0.60 | 262,144 | — | source 2026-11-14 |
| Qwen3-VL-32B-Thinking | standard | USD 0.20 | — | USD 1.50 | 262,144 | — | source 2026-11-14 |
| Qwen3.5 122B A10B | standard | USD 0.26 | — | USD 2.08 | 262,144 | −35% | source 2026-11-14 |
| Qwen3.5 27B | standard | USD 0.25 | — | USD 2 | 262,144 | −17% | source 2026-11-14 |
| Qwen3.5 35B A3B | standard | USD 0.24 | USD 0.15 | USD 1.80 | 262,144 | −8% | source 2026-11-14 |
| Qwen3.5 9B | standard | USD 0.10 | — | USD 0.15 | 262,144 | — | source 2026-11-14 |
| Qwen3.6 27B | standard | USD 0.30 | — | USD 3.20 | 262,144 | −24% | source 2026-11-14 |
| Qwen3.6 35B A3B | standard | USD 0.24 | USD 0.15 | USD 1.80 | 262,144 | −25% | source 2026-11-14 |
| Qwen3.8 2.4T A95B | standard | USD 2 | USD 0.25 | USD 6 | 1,048,576 | same | source 2026-11-14 |
| Seed-OSS-36B-Instruct | standard | USD 0.21 | — | USD 0.57 | 262,144 | — | source 2026-11-14 |
| Step-3.5-Flash | standard | USD 0.10 | — | USD 0.30 | 262,144 | — | source 2026-11-14 |
- Cached "—" means SiliconFlow's page shows no cache price for that row; budget cached tokens at the input price.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What service tiers does SiliconFlow publish?
Standard prices only: SiliconFlow publishes no per-model batch, flex, priority or fast price on the pages we read.
Cached input: 26 of 56 standard prices include a cache-read price, a median 20% of the input price. What cached input means.
What we recorded about SiliconFlow's pricing page: International site (siliconflow.com), USD. Read from the Serverless Pricing tables after expanding every "Load More". Context lengths are shown rounded (e.g. 262K) and converted to the usual token counts. Model IDs are not shown on the pricing page. Image, video and audio models excluded.
Is SiliconFlow cheaper than the model makers' own APIs?
Mostly yes. Of 19 standard prices for models whose maker sells its own API, 14 are below the maker's price, 0 above and 5 the same, on the 3:1 blend. The median difference is 10% below the maker. The biggest saving is DeepSeek V4.1 Flash, 50% below DeepSeek.
BroadNet Inference and SiliconFlow
BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 3 models that SiliconFlow prices: BroadNet Gemma 4 VL 26B at USD 0.10 / USD 0.22 against SiliconFlow's USD 0.14 / USD 0.40 (BroadNet is cheaper on the 3:1 blend); DeepSeek V4.1 Flash at USD 0.175 / USD 0.52 against SiliconFlow's USD 0.15 / USD 0.60 (BroadNet is cheaper on the 3:1 blend); GLM 5.3 at USD 0.78 / USD 3.46 against SiliconFlow's USD 1.40 / USD 4.40 (BroadNet is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.
More on SiliconFlow
- SiliconFlow's own pricing page (the source; checked 2026-11-14)
- All 41 providers compared · Cheapest LLM APIs right now · Download every price (CSV, JSON)
Questions
How much does the SiliconFlow API cost per token?
SiliconFlow's standard prices run from USD 0.05 input and USD 0.05 output per 1M tokens (Qwen 2.5 7B) to USD 2.70 / USD 13.50 (Kimi K3). The median across 56 standard prices is USD 0.25 input and USD 1 output per 1M, which is USD 0.00000025 per input token. Input, cached input and output are billed separately.
What is the cheapest model on SiliconFlow?
Qwen 2.5 7B, at USD 0.05 input and USD 0.05 output per 1M tokens (USD 0.05 on a 3:1 input-to-output blend). The cheapest from other makers: DeepSeek V4 Flash (DeepSeek) at USD 0.13 / USD 0.28; GLM 5.3 Flash (Z.ai) at USD 0.15 / USD 0.50; Kimi K2.5 (Moonshot AI) at USD 0.45 / USD 2.25.
Is SiliconFlow cheaper than other providers?
It depends on the model. Of 47 models that other providers also price, SiliconFlow is the cheapest standard price for 11, for example Kimi K2.5, Qwen3 Coder 480B A35B and Qwen3.5 122B A10B; it ties the lowest price, without beating it, for 6 more; another provider is cheaper for 30; and it is the dearest for 4. Its median rank is 2 of a median 7 providers, on a 3:1 input-to-output blend.
Does SiliconFlow have batch, cached or priority pricing?
Cached input: yes, for 26 of 56 standard prices, at a median 20% of the input price. No batch, flex or priority prices are listed per model on the pages we read.
How much does a chatbot cost per month on SiliconFlow?
For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 5.40 on Qwen 2.5 7B, USD 40.50 on Qwen3 Coder 480B A35B (the middle of SiliconFlow's list) and USD 354.78 on Kimi K3. Try your own numbers.