Catalogue
Models
455 models, and deliberately not every model there is: one earns a place by being sold by a provider we track, measured by a leaderboard we follow, or published as open weights by a lab we follow. Pointer names like “-latest” stay out, because they answer for whatever the vendor aimed them at that week.
How we choose what to list3 models
KAT-Coder-Air V2.5KwaipilotKAT-Coder-Air V2.5 is Kwaipilot's hosted-only coding assistant built for long-document code review and generation.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.15 / $0.60per 1Mvia StreamLake
KAT-Coder-Pro2 sizes here
KAT-Coder-Pro V2.5KwaipilotKAT-Coder-Pro is a proprietary coding model from Kwaipilot built for long-context code tasks, with identical pricing across its three providers and no published benchmark scores in our data. Its 256,000-token request limit suits large-repository work, though you must choose between them on speed rather than cost.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.74 / $2.96per 1Mvia OpenRouterKAT-Coder-Pro V2KwaipilotKAT-Coder-Pro V2 is Kwaipilot's proprietary coding model built for long-context code tasks, handling up to 262,144 tokens in a single request.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.30 / $1.20per 1Mvia OpenRouter