Skip to main content
DeepSeek-V4-Flash-0731 is live — read the docsUNIVERSAL GATEWAY

Every model. Every modality.One ModelSite endpoint.

OpenAI & Anthropic protocols · chat · video · audio · embeddings · rerank — one key to all of it.

Create a free API keyBrowse models & prices
$curl -s https://api.modelsite.ai/v1/chat/completions-H "Authorization: Bearer <MODELSITE_API_KEY>"
Models
29+
New models added continuously
Vendors
8+
Direct vendor channels
Modalities
5 modalities
chat · video · audio · embeddings · rerank
Live Routing Topology

One ingress. Lossless streaming.

Any OpenAI-compatible client enters through a single gateway; the routing core forwards losslessly to upstream clusters.

Source Ingress
Unified client ingress
OpenAI / Anthropic SDKs, LangChain, plain REST — swap one line of base_url and you are in.
Port 443 · TLS 1.3HTTP/2
⚡ Modelsite CoreMulti-vendor routing
Routing core
Requests are dispatched to upstream channels by model and load; degraded channels are bypassed automatically.
One bill · one monthly statementOpenAI & Anthropic
Upstream NodesDirect
Upstream clusters
GLM, DeepSeek, Kimi, Anthropic and more — catalog and prices stay in sync.
Only financial summaries storedChat content never persisted
Streaming passthrough · no prompt persistence · encrypted end to end
Billing: list price × your discount
Illustrative diagram — not an availability promise
Curated Registry

Curated frontier showcase

Change one request parameter, model: "..." for an instant switch.

Browse all models
Zhipu
Best value
GLM-5.3
zhipu · bigmodel

A flagship with strong code and reasoning alignment, at a fraction of closed-model pricing.

Price (1M)$0.80 / $2.90
P90 latency~ 260ms
DeepSeek
Deep reasoning
DeepSeek-V4
deepseek

Long-thinking MoE flagship pushing the frontier on math and code.

Price (1M)$0.55 / $2.20
P90 latency~ 300ms
Qwen
All-round flagship
Qwen-3.7
alibaba / qwen

A balanced flagship strong across long context and tool use.

Price (1M)$0.36 / $1.40
P90 latency~ 320ms
MoonshotAI
Open source
Kimi K2.7
moonshot

High-throughput open-weight flagship, friendly to agentic tooling and batch workloads.

Price (1M)$0.60 / $2.50
P90 latency~ 240ms
Savings Calculator

Estimate your monthly enterprise savings

Model
Monthly API calls5,000,000
Tokens per call2,000
Your discount saves monthly⚡ Real discount
Sign in to see

Prices come live from the catalog; assumes a 1:3 input/output ratio.

List-price monthly spend$23,750.00
Your discounted spendSign in to see
Unlock your discount now

Ready to unify your model access?

Swap one line of base_url and call every model with a single key. Transparent pricing, pay as you go.

Create a free API keyRead the API docs