The table below is the same view SchneeAI’s router uses internally — minus the per-tenant policy overlays. Filter by capability, sort by effective cost or context window, and watch how the price column updates as you toggle. Everything here is client-side.
How to read the price
The bold number in each row is what you pay through SchneeAI: provider list price × ×1.5 platform factor. The smaller reference row below is the raw provider list price. For the factor breakdown and when it is worth paying, see the Cost Calculator and its TCO section.
Filters and sort
- Capability filters — narrow the table to models that support vision, tool use, reasoning, or that sit in the “cheap” tier (roughly: under $1/1M input).
- Sort — by total cost ascending (cheapest first), descending, context size, provider, or model name.
- Latency — our operational bucketing, not provider SLAs. “Fast” targets sub-second first token; “Medium” targets 1-3s; “Slow” is typical for deep-reasoning models that emit chain-of-thought before answering.
What is not here
We deliberately omit quality rankings, benchmark scores, and “best for X” claims. Those move every release and depend heavily on your prompts and eval harness. The honest answer is: try the model on your workload, measure, then route. SchneeAI’s model: "auto" mode does exactly this with the routing policy you configure per feature.
Adding a model
If a model you need is missing from this list, it does not mean SchneeAI cannot route to it — LiteLLM supports 100+ providers and we expose new models per tenant request. Email [email protected] with the provider and model name, and we will enable it on your workspace.