PPIO Model API Service lets you quickly access and use industry-leading open-source models, including LLMs, image models, audio models, and video models. You pay only for usage.
Limited-time 50% off input and output tokens with batch inference (select models supported)
Learn moreDeepSeek
Advanced AI models from Deepseek offering cutting-edge reasoning capabilities and highly competitive pricing for enterprise and research applications
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | - | 1M | ¥1 | - | ¥0.02 | ¥4 | |
| DeepSeek V4 Flash Vision Exp | - | 1M | ¥3 | - | ¥1 | ¥9 | |
| DeepSeek V4 Pro 0813 | - | 1M | ¥9 | - | ¥3 | ¥27 | |
| DeepSeek V4 Flash 0731 | - | 1M | ¥3 | - | ¥1 | ¥9 | |
| DeepSeek V4 Pro | - | 1M | ¥9 | - | ¥3 | ¥27 | |
| DeepSeek OCR 2 | - | 8.2K | ¥0.216 | - | - | ¥0.216 |
Zhipu
Zhipu series language models
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| GLM 5.3 FlashX | - | 1M | ¥2 | - | ¥0.57 | ¥7 | |
| GLM 5.3 Flash | - | 1M | ¥0.8 | - | ¥0.23 | ¥2.8 | |
| GLM 5.3 | - | 1M | ¥8 | - | ¥2 | ¥28 | |
| GLM 5.2 | - | 1M | ¥8 | - | ¥2 | ¥28 | |
| GLM 5.1 | 1–32.768K | 204.8K | ¥6 | - | ¥1.3 | ¥24 | |
| 32.768K–204.8K | 204.8K | ¥8 | - | ¥2 | ¥28 | ||
| GLM 5 | 1–32K | 202.8K | ¥4 | - | ¥1 | ¥18 | |
| 32K–204.8K | 202.8K | ¥6 | - | ¥1.5 | ¥22 | ||
| GLM 4.7 Flash | - | 200K | ¥0.5 | - | ¥0.1 | ¥3 | |
| GLM 4.7 | 1–32K | 204.8K | ¥2 | - | ¥0.4 | ¥8 | |
| 1–32K | 204.8K | ¥3 | - | ¥0.6 | ¥14 | ||
| 32K–204.8K | 204.8K | ¥4 | - | ¥0.8 | ¥16 | ||
| AutoGLM Phone 9B Multilingual | - | 65.5K | ¥0.25 | - | - | ¥1 | |
| GLM 4.6v | 1–32.768K | 131.1K | ¥1 | - | - | ¥3 | |
| 32.768K–131.072K | 131.1K | ¥2 | - | - | ¥6 | ||
| GLM 4.6 | - | 204.8K | ¥4 | ¥0.8(5m) | ¥0.8 | ¥16 | |
| GLM 4.5V | - | 65.5K | ¥4 | - | ¥0.8 | ¥12 | |
| GLM 4.5 Air | - | 131.1K | ¥1.2 | - | ¥0.24 | ¥2 |
Qwen
Qwen series models provide efficient language processing across parameter sizes, from lightweight to enterprise-grade solutions
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| Qwen3.8 Flash | - | 1M | ¥0.8 | ¥1.25(5m) | ¥0.1 | ¥2.7 | |
| Qwen3.8 27B | - | 1M | ¥3 | - | ¥0.6 | ¥12 | |
| Qwen3.8 2.4T A95B | - | 1M | ¥12 | - | ¥1.5 | ¥36 | |
| Qwen3.8 Max | - | 1M | ¥12 | ¥15(5m) | ¥1.5 | ¥36 | |
| Qwen3.7 Max | - | 1M | ¥6 | - | ¥1.2 | ¥18 | |
| Qwen3.6 27B | - | 262.1K | ¥3 | - | - | ¥18 | |
| Qwen3.5 27B | 1–128K | 262.1K | ¥0.6 | - | - | ¥4.8 | |
| 128K–262.144K | 262.1K | ¥1.8 | - | - | ¥14.4 | ||
| Qwen3.5 122B A10B | 1–128K | 262.1K | ¥0.8 | - | - | ¥6.4 | |
| 128K–262.144K | 262.1K | ¥2 | - | - | ¥16 | ||
| Qwen3.5 35B A3B | 1–128K | 262.1K | ¥0.4 | - | - | ¥3.2 | |
| 128K–262.144K | 262.1K | ¥1.6 | - | - | ¥12.8 | ||
| Qwen3.5 397B A17B | 1–128K | 262.1K | ¥1.2 | - | - | ¥7.2 | |
| 128K–262.144K | 262.1K | ¥3 | - | - | ¥18 | ||
| Qwen3.6 35B A3B | - | 262.1K | ¥1.8 | - | - | ¥10.8 |
MiniMax
MiniMax series language models
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| MiniMax M3.1 Flash | - | 1M | ¥1 | - | ¥0.1 | ¥4 | |
| MiniMax-M3 | 1–524.288K | 1M | ¥2.1 | - | ¥0.42 | ¥8.4 | |
| 524.288K–1000K | 1M | ¥4.2 | - | ¥0.84 | ¥16.8 | ||
| MiniMax M2.7-highspeed | - | 204.8K | ¥4.2 | ¥2.625(5m) | ¥0.42 | ¥16.8 | |
| MiniMax M2.7 | - | 204.8K | ¥2.1 | ¥2.625(5m) | ¥0.42 | ¥8.4 | |
| MiniMax M2.5 | - | 204.8K | ¥2.1 | ¥2.625(5m) | ¥0.21 | ¥8.4 | |
| MiniMaxAI/MiniMax-M1-80k | - | 128K | ¥4 | - | - | ¥16 |
Moonshot AI
Moonshot AI series language models
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| Kimi K3.1 Pro Highspeed | - | 1M | ¥26 | ¥26(5m)·¥52(1h) | ¥1.3 | ¥136 | |
| Kimi K3.1 Pro | - | 1M | ¥13 | ¥13(5m)·¥26(1h) | ¥1.3 | ¥68 | |
| Kimi K3 | - | 1M | ¥20 | ¥20(5m)·¥40(1h) | ¥2 | ¥100 | |
| Kimi K2.7 Code | - | 262.1K | ¥6.5 | - | ¥1.3 | ¥27 | |
| Kimi K2.6 | - | 262.1K | ¥6.5 | - | ¥1.1 | ¥27 | |
| Kimi K2.5 | - | 262.1K | ¥4 | - | ¥0.7 | ¥21 | |
| Kimi K2 Thinking | - | 262.1K | ¥4 | - | - | ¥16 | |
| Kimi K2 Instruct 0905 | - | 262.1K | ¥4 | - | - | ¥16 | |
| Kimi K2 Instruct | - | 131.1K | ¥4 | - | - | ¥16 |
inclusionAI
inclusionAI series language models
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| Ling 3.1 Flash 1019 | - | 1M | ¥2 | - | ¥0.4 | ¥6 |
StepFun
StepFun series language models
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| Step 5 Preview | - | 1M | ¥7 | - | ¥0.7 | ¥20 |
Other models
Other models series language models
| Model name | Tiered | Context | Input (/Mt) | Cache write (/Mt) | Cache read (/Mt) | Output (/Mt) | Action |
|---|---|---|---|---|---|---|---|
| PPIO 4B | - | 128K | ¥0.2145 | - | - | ¥0.2145 |
| Model name | Context length | Price |
|---|---|---|
| Qwen3 Embedding 8B | 32768 | ¥0.35/million Tokens |
| BAAI:BGE-M3 | 8192 | ¥0.07/million Tokens |
More services
Request a quote
Our service team is always online to provide continuous support and peace of mind
Contact usResource Partnership
We invite partners to precisely match needs and explore new business opportunities together
Contact us