Open-weight models built for agentic work.Fast first token, high throughput, low prices.
A short list of the best open models for coding agents and tool use: low time to first token, high tokens per second, unquantized BF16 weights where available, and pay-as-you-go prices, or a monthly Agent plan. Add funds or pick a plan, get an API key, and pay only for what you use.
What is Tokener CA?
Tokener CA is an API for open-weight AI models. Create an API key and call the models from your own code or from the tools you already use, then pay the way that suits you: as you go, with funds you add to your balance, or with a monthly Agent plan that includes usage up to a 5-hour and a weekly limit.
Instead of listing hundreds of models, Tokener CA offers a short, curated catalog of open models that do well at agentic work, with a fast first token, high throughput and low prices. Each model's price, context window, precision and capabilities are published on the models page.
One API key works with three protocols: OpenAI-compatible (Chat Completions), OpenAI Responses and Anthropic-compatible (Messages). Code written for the OpenAI or Anthropic SDKs works after changing the base URL, the API key and the model id. Tokener CA is operated by ShowKey Information Systems Inc., a Canadian company based in the Greater Toronto Area, Ontario.
Featured models
A few of the open models available right now
Precision — BF16 means 16-bit weights that are not quantized: the model as its maker released it. FP8 means 8-bit weights, often the version the model's maker publishes; quality stays close to 16-bit, and it runs faster. What precision means
Plans
Pay as you go at per-token prices, or a monthly Agent plan that covers your usage up to a 5-hour and a weekly limit — on the same key and the same endpoints.
Pay as you go
For apps, APIs and everything else
- Any amount from US$2.00
- Pay per token at the published prices
- Unused funds stay on your balance