Explore pricing for our Model APIs and GPU resources. Find the right plan to match your needs with transparent rates and flexible options.
Batch inference is available at an introductory 50% discount on input and output tokens for supported models. Learn More
Modelos avançados de IA da DeepSeek, oferecendo recursos de raciocínio de ponta e preços competitivos para aplicações empresariais e de pesquisa.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | - | 1M | $0.3 | - | $0.006 | $1.2 | More |
| DeepSeek V4.1 Flash | - | 1M | $0.3 | - | $0.006 | $1.2 | More |
| DeepSeek V4 Pro 0813 | - | 1M | $1.32 | - | $0.044 | $3.96 | More |
| DeepSeek V4 Flash 0731 | - | 1M | $0.44 | - | $0.028 | $1.32 | More |
| DeepSeek V4 Pro | - | 1M | $1.6 | - | $0.135 | $3.2 | More |
| DeepSeek OCR 2 | - | 8K | $0.03 | - | - | $0.03 | More |
| DeepSeek R1 0528 | - | 164K | $0.7 | - | $0.35 | $2.5 | More |
| DeepSeek R1 Turbo | - | 64K | $0.7 | - | - | $2.5 | More |
Modelos da série GLM da Universidade Tsinghua, com compreensão avançada da língua chinesa e recursos de geração.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| GLM 5.3 FlashX | - | 1M | $0.37 | - | $0.075 | $1.25 | More |
| GLM 5.3 Flash | - | 1M | $0.15 | - | $0.03 | $0.5 | More |
| GLM 5.3 | - | 1M | $1.4 | - | $0.26 | $4.4 | More |
| GLM 5.2 | - | 1M | $1.4 | - | $0.26 | $4.4 | More |
| GLM 5.1 | - | 205K | $1.38 | - | $0.26 | $4.4 | More |
| GLM 5 | - | 203K | $1 | - | $0.2 | $3.2 | More |
| GLM 4.7 Flash | - | 200K | $0.07 | - | $0.01 | $0.4 | More |
| GLM 4.7 | - | 205K | $0.6 | - | $0.11 | $2.2 | More |
| AutoGLM Phone 9B Multilingual | - | 66K | $0.035 | - | - | $0.138 | More |
| GLM 4.6V | - | 131K | $0.3 | - | $0.055 | $0.9 | More |
| GLM 4.6 | - | 205K | $0.55 | - | $0.11 | $2.2 | More |
| GLM 4.5V | - | 66K | $0.6 | - | $0.11 | $1.8 | More |
| GLM 4.5 Air | - | 131K | $0.13 | - | $0.025 | $0.85 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Kimi K3.1 Pro Highspeed | - | 1M | $4 | $4(5m)$8(1h) | $0.2 | $20 | More |
| Kimi K3.1 Pro | - | 1M | $2 | $2(5m)$4(1h) | $0.2 | $10 | More |
| Kimi K3 | - | 1M | $3 | $3(5m)$6(1h) | $0.3 | $15 | More |
| Kimi K2.7 Code | - | 262K | $0.95 | - | $0.19 | $4 | More |
| Kimi K2.6 | - | 262K | $0.8 | - | $0.16 | $3.4 | More |
| Kimi K2.5 | - | 262K | $0.6 | - | $0.1 | $3 | More |
| Kimi K2 Thinking | - | 262K | $0.6 | - | $0.15 | $2.5 | More |
| Kimi K2 Instruct 0905 | - | 262K | $0.6 | - | - | $2.5 | More |
| Kimi K2 Instruct | - | 131K | $0.57 | - | - | $2.3 | More |
Modelos Llama da Meta que oferecem compreensão de linguagem de última geração com arquitetura aberta, projetada para diversas aplicações.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Llama 3.1 8B Instruct | - | 16K | $0.02 | - | - | $0.05 | More |
| Llama 3.3 70B Instruct | - | 12K | $0.135 | - | - | $0.4 | More |
| Llama 4 Scout Instruct | - | 131K | $0.18 | - | - | $0.59 | More |
Modelos da série Qwen que oferecem processamento de linguagem eficiente com vários tamanhos de parâmetros, desde soluções leves até soluções de nível empresarial.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Qwen3.8 Flash | - | 1M | $0.15 | $0.2(5m) | $0.016 | $0.47 | More |
| Qwen3 VL Plus | [1, 32K) | 262K | $0.144 | $0.18(5m) | $0.014 | $1.434 | More |
| [32K, 128K) | 262K | $0.216 | $0.27(5m) | $0.022 | $2.151 | ||
| [128K, 256K) | 262K | $0.431 | $0.539(5m) | $0.043 | $4.301 | ||
| Qwen4 Max | - | 1M | $1.413 | $1.767(5m)$2.827(1h) | $0.071 | $4.24 | More |
| Qwen3.8 27B | - | 1M | $0.42 | - | $0.085 | $3 | More |
| Qwen3.8 2.4T A95B | - | 1M | $2 | - | $0.25 | $6 | More |
| Qwen3.8 Max | - | 1M | $2 | $2.5(5m) | $0.25 | $6 | More |
| Qwen3.7 Max | - | 1M | $1.25 | - | $0.25 | $3.75 | More |
| Qwen3.6 27B | - | 262K | $0.6 | - | - | $3.6 | More |
| Qwen3.5 27B | - | 262K | $0.3 | - | - | $2.4 | More |
| Qwen3.5 122B A10B | - | 262K | $0.4 | - | - | $3.2 | More |
| Qwen3.5 35B A3B | - | 262K | $0.25 | - | - | $2 | More |
| Qwen3.5 397B A17B | - | 262K | $0.6 | - | - | $3.6 | More |
| Qwen3.6 35B A3B | - | 262K | $0.248 | - | - | $1.485 | More |
| Qwen3 VL 30B A3B Instruct | - | 131K | $0.2 | - | - | $0.7 | More |
| Qwen3 Omni 30B A3B Thinking | - | 66K | $0.25(text)$2.2(audio)$0.46(image)$0.46(video)$0.46(image/video) | - | - | $0.97(text)$1.788(audio) | More |
| Qwen MT Plus | - | 16K | $0.25 | - | - | $0.75 | More |
Modelos Gemma do Google que oferecem processamento de linguagem de alta qualidade com excelente desempenho para diversas tarefas de NLP.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Gemma 4 26B A4B | - | 262K | $0.13 | - | - | $0.4 | More |
| Gemma 4 31B | - | 262K | $0.14 | - | - | $0.4 | More |
| Gemma 3 27B | - | 98K | $0.119 | - | - | $0.2 | More |
Os modelos da Microsoft oferecem recursos multilíngues de última geração com segurança de nível empresarial, perfeitamente integrados ao ecossistema de nuvem Azure. Otimizados para colaboração multiplataforma e inteligência de negócios, esses modelos se destacam na compreensão e geração de documentos — pontos fortes essenciais para fluxos de trabalho do Microsoft 365.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| WizardLM 2 8x22B | - | 66K | $0.62 | - | - | $0.62 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Ling 3.1 Flash 1019 | - | 1M | $0.3 | - | $0.06 | $0.9 | More |
| Ling 3.0 Flash Fin | - | 262K | $0.075 | - | $0.015 | $0.22 | More |
| Ling 3.0 Flash | - | 262K | $0.06 | - | $0.012 | $0.18 | More |
| Ling 3.0 Flash VL | - | 262K | $0.075 | - | $0.015 | $0.22 | More |
| Ling 3.0 Flash Sante | - | 262K | $0.075 | - | $0.015 | $0.22 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Macaron V1 Venti | - | 1M | $0.825 | - | $0.165 | $2.475 | More |
| Macaron V1 Tall | - | 262K | $0.2475 | - | $0.044 | $1.43 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| NVIDIA Nemotron 3 Nano 30B A3B | - | 262K | $0.05 | - | - | $0.2 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| OpenAI GPT OSS 120B | - | 131K | $0.05 | - | - | $0.25 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Step 5 Preview | - | 1M | $1 | - | $0.1 | $2.7 | More |
| Step 3.7 Flash | - | 262K | $0.2 | - | $0.04 | $1.15 | More |
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| MiMo V2.6 Flash | - | 1M | $0.14 | - | $0.0028 | $0.28 | More |
| MiMo V2.6 Pro | - | 1M | $0.435 | - | $0.0036 | $0.87 | More |
Modelos especializados ajustados finamente, otimizados para aplicações criativas e de roleplay, com capacidades aprimoradas de narrativa.
| Model | Tier | Context | Input(/Mt) | Cache Write(/Mt) | Cache Read(/Mt) | Output(/Mt) | Actions |
|---|---|---|---|---|---|---|---|
| Llama 3 8B Stheno V3.2 | - | 8K | $0.05 | - | - | $0.05 | More |
| Model Name | Context | Price |
|---|---|---|
| Qwen3 Embedding 8B | 33K | $0.07/Mt |
| BAAI:BGE-M3 | 8K | $0.01/Mt |
| API Name | Mode | Pricing |
|---|---|---|
| EXA | neuralSearch | $0.007/request |
| deepSearch | $0.012/request | |
| deepReasoningSearch | $0.015/request | |
| additional_result | $0.001/result (beyond 10 results) | |
| answer | $0.005/request | |
| contentText | $0.001/item | |
| contentHighlight | $0.001/item | |
| contentSummary | $0.001/item | |
| Tavily | basicSearch | $0.008/request |
| advancedSearch | $0.016/request | |
| basicExtract | $0.0016/url | |
| advancedExtract | $0.0032/url | |
| regularMapping | $0.0008/page | |
| instructedMapping | $0.0016/page | |
| Crawl | Extract + Mapping Cost |
200+ models, on-demand GPUs, and secure agent runtimes — unified under one API. Free to start, scales as you grow.