ENTERPRISE MODEL & COMPUTE CLOUD LIVE
Call models, deploy services,
and rent GPUs — in one place
O17Pod provides DeepSeek V4 Pro, DeepSeek V4 Flash, Kimi K3 and other leading model services through one API or dedicated inference deployments on Kubernetes and Docker hosts.
✓ DeepSeek V4 / Kimi K3✓ OpenAI compatible✓ Per-key billing controls
O17Pod COMMAND CENTERTenant: Production ● Healthy
PRODUCTION WORKSPACEGood evening, BuilderLIVE
Token balance86.4M↑ 12.8%
Instances12/ 50 quota
API calls today1.28M↑ 18.6%
API traffic
RUNNINGDeepSeek-R1-32B4 × A100128 ms
RUNNINGQwen2.5-72B2 × L40S96 ms
Request routing12.8K RPS+24.6%
01 / PRODUCT CAPABILITIES
From model calls to compute delivery
O17Pod combines an AI gateway, inference deployment, GPU scheduling and commercial billing in one MaaS platform.
⌘01
Unified Model API
One key and one endpoint across providers.
- OpenAI & Anthropic compatibility
- Streaming and tool use
- Routing, fallback and retries
⬡02
Dedicated Deployment
Turn open or fine-tuned models into private APIs.
- vLLM, TGI and SGLang
- Kubernetes and Docker
- Autoscaling, canary and rollback
▱03
GPU Compute Rental
Launch training, inference and notebook containers.
- H100, A100, L40S and 4090
- Custom images and storage
- Hourly, daily and monthly billing
◉04
Tokens & Billing
Control prepaid and usage-based spend.
- Token plans and PAYG
- Tenant, project and key budgets
- Orders, invoices and alerts
02 / MODEL CATALOG
DeepSeek, Kimi, MiniMax and other leading models
O17Pod provides unified model API and dedicated enterprise deployment services, with model choice based on capability, latency, cost and compliance.
Operational
DAVAILABLE
DeepSeek V4 Pro
DeepSeek · Reasoning / AgentsO17Pod model serviceIntegrated
DAVAILABLE
DeepSeek V4 Flash
DeepSeek · Chat / ReasoningO17Pod model serviceIntegrated
KAVAILABLE
Kimi K3
Moonshot AI · Multimodal / ReasoningO17Pod model serviceIntegrated
MAVAILABLE
MiniMax M3
MiniMax · Multimodal / AgentsO17Pod model serviceIntegrated
HAVAILABLE
MiniMax H3
MiniMax · Video / Omni-generationO17Pod model serviceIntegrated
DAVAILABLE
DeepSeek R1
DeepSeek · ReasoningO17Pod model serviceIntegrated
QAVAILABLE
Qwen 2.5 72B
Alibaba · ChatO17Pod model serviceIntegrated
LAVAILABLE
Llama 3.3 70B
Meta · ChatO17Pod model serviceIntegrated
VAVAILABLE
Qwen2-VL 72B
Alibaba · MultimodalO17Pod model serviceIntegrated
BAVAILABLE
BGE-M3
BAAI · EmbeddingO17Pod model serviceIntegrated
SAVAILABLE
Stable Diffusion XL
Stability AI · ImageO17Pod model serviceIntegrated
03 / COMPUTE RENTALTwo delivery modes,
⬡Kubernetes PodElastic scheduling · self-healing · per-second billingRECOMMENDED ▱Docker on selected hostsFixed node · port mapping · container lifecycle
Two delivery modes,
one compute experience
Launch cloud-native Kubernetes Pods or rent containers on designated Docker GPU hosts with unified inventory, images, networking, monitoring and billing.
04 / DEVELOPER EXPERIENCEChange one URL.
Change one URL.
Keep your code.
Keep your OpenAI SDK and application logic. Replace the API key, base URL and model name.
- ● Chat Completions
- ● Responses API
- ● Anthropic Messages
- ● Embedding & Rerank
quickstart.py
from openai import OpenAI
client = OpenAI(
api_key="YOUR_O17POD_API_KEY",
base_url="https://api.o17.dev/v1"
)
response = client.chat.completions.create(
model="deepseek-r1",
messages=[{"role":"user","content":"Hello O17Pod"}],
stream=True
)05 / ENTERPRISE PLATFORM
Built for production
Multi-tenant isolation, RBAC, encryption, audit trails, private networking and full-stack observability.
Web AppAI AgentEnterprise RAGOpenAI SDK
↓Multi-Protocol API GatewayAUTHRATE LIMITROUTINGMETERING
↓06 / PRICING
Scale from prototype to production
Choose public models, token plans, private deployments and GPU compute as your workload grows.
Developer
¥0
- ✓ Public model API
- ✓ Starter token plans
- ✓ 20 RPM default
- ✓ Community support
Team
¥1,680 / month
- ✓ High-concurrency API
- ✓ Projects and members
- ✓ Budgets, quotas and alerts
- ✓ GPU rental discounts
Enterprise
Custom
- ✓ Dedicated deployments
- ✓ Reserved GPU pools
- ✓ Enterprise SLA and support
- ✓ Private and hybrid cloud
07 / OPEN ECOSYSTEM
Built on proven open source
O17Pod adds one product, billing and operations layer over dependable infrastructure.
Kubernetes
Docker
vLLMvLLM
SGLangSGLang
Hugging Face
Prometheus
Grafana
OpenTelemetry
READY TO BUILD?From your first API call
From your first API call
to production-scale inference
Manage models, compute, tokens and spend in one place — and focus on your AI product.