One API for
every AI model
DeepSeek, GLM, Kimi, Qwen, Llama, Gemma and more, behind a single key. Prepaid credits, per-token prices you can see, and no subscriptions.
# One key. Pick any model by name. curl https://solorouter.ai/api/v1/chat/completions \ -H "Authorization: Bearer $SOLO_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek/deepseek-v4.1-flash", "messages": [ {"role": "user", "content": "Hello"} ] }' # Switch model: "z-ai/glm-5.3", "moonshotai/kimi-k3" ...
Built for developers who ship
Everything you need to use the best AI models in production, without managing a dozen provider accounts.
One key for everything
Switch models by changing one string. Uses the standard chat-completions format, so LangChain, Cline, Continue and most AI tools work by changing one URL.
Prepaid, no surprises
Top up from $5. Every response reports its exact cost, and each key can have its own spending limit.
Images, video and voice
FLUX, Stable Diffusion, Wan 2.2 video, ElevenLabs and Qwen speech, Stable Audio music, plus embeddings and reranking. Same key, same balance.
Streaming and tools
Server-sent events, function calling, JSON mode and reasoning output, passed straight through.
Prompts stay private
We log tokens and cost for your bill, never the content of your prompts or responses.
Chat in the browser
Not writing code? Sign in and chat with any model, billed from the same balance.
Transparent per-token pricing
USD per million tokens.
| Model | Provider | Context | Input | Output |
|---|---|---|---|---|
DeepSeek V4.1 Flash deepseek/deepseek-v4.1-flash | DeepSeek | 1M | $0.39 | $1.56 |
|
DeepSeek V4 Pro deepseek/deepseek-v4-pro | DeepSeek | 1M | $2.26 | $4.52 |
GLM-5.3 z-ai/glm-5.3 | Z.ai | 1M | $1.82 | $5.72 |
|
GLM-5.3 Flash z-ai/glm-5.3-flash | Z.ai | 1M | $0.20 | $0.65 |
Kimi K3 moonshotai/kimi-k3 | Moonshot AI | 262K | $3.90 | $19.50 |
Qwen 3.8 Max qwen/qwen3.8-max | Alibaba | 1M | $2.60 | $7.80 |
Gemma 4 31B google/gemma-4-31b-it | 256K | $0.23 | $0.65 | |
Llama 4 Maverick meta-llama/llama-4-maverick | Meta | 128K | $0.33 | $1.13 |
Every major open model, one bill
Pick a maker to see its models, prices and a ready-to-run example.
From sign-up to first response in three steps
Sign in with Google
Your account comes with free credit and a ready-made API key. No forms, no sales call.
Point your code at us
Set the base URL to https://solorouter.ai/api/v1 and use your key. Existing chat-completions code keeps working.
Pick any model
Change the model string to switch between DeepSeek, GLM, Kimi, Qwen and more. Top up from $5 when you need more.
What teams build with Solo Router
Coding agents
Run Cline, Continue or your own agent on GLM-5.3, Kimi K2.6 or DeepSeek V4 Pro, with 1M-token context on many models and cached input to keep long sessions cheap.
Chatbots and assistants
Ship a support bot on a fast, low-cost model like GLM-5.3 Flash or DeepSeek V4 Flash, and route hard questions to a stronger model.
Search and RAG
Embed documents with BGE-M3 or E5, then answer questions over them with any chat model, all on the same key and balance.
Images
Generate product shots, thumbnails and illustrations with Stable Diffusion 3.5 Large for a fixed price per image.
Drop-in for the tools you already use
Anything that speaks the standard chat-completions format works by setting a base URL and key.
Questions, answered
What is Solo Router?
Solo Router is an AI model router: one API key and one prepaid balance for models from DeepSeek, Z.ai (GLM), Moonshot (Kimi), Alibaba (Qwen), Meta, Google, MiniMax, NVIDIA and more.
How much does it cost?
You pay per token at each model's listed price, from prepaid credit. Top up from $5; new accounts get free credit. There is no subscription.
Do I need to change my code?
Usually not. Solo Router uses the standard chat-completions request format, so you only change the base URL to https://solorouter.ai/api/v1 and use your Solo Router key.
Do you store my prompts?
No. We log token counts and cost for billing, never the content of your prompts or responses.
Can I limit spending?
Yes. Requests stop when your balance reaches zero, and each API key can have its own spending limit.
Which models do you offer?
Chat and reasoning models such as DeepSeek V4.1, GLM-5.3, Kimi K3, Qwen 3.8 Max, Llama 4 and Gemma 4, plus Stable Diffusion 3.5 for images and embedding models for search.
Guides and pricing breakdowns
The cheapest LLM APIs right now (price per million tokens)
The cheapest LLM APIs on Solo Router ranked by price per million tokens, with advice on when a cheap model is good enough and when to pay for a bigger one.
DeepSeek API pricing explained (with real cost examples)
How DeepSeek API pricing works per million tokens, what cached tokens and reasoning tokens cost, and what a typical chat or coding request costs on Solo Router.
Free LLM API options in 2026 (and their limits)
Where to get a free LLM API in 2026: free credits, free tiers and free models, plus the catches (rate limits, data use, model choice) and when paying a little is better.
Start building in minutes
Sign in with Google and get free credit to try any model.
Create your API key