ModelBench
Model catalog

Models

Every model in the catalog with source-linked pricing, context limits, provider availability, and published benchmark results.

Compare Export CSV
3 models
Moonshot AI jiekou Open weights
Moonshot AI: Kimi K2.5 T
Reasoning Tools JSON Open weights

Earlier Kimi frontier model for long-context agents, coding, and multimodal work

moonshotai/kimi-k2.5 2026-01-27 262.144K context $0.375/M input $2/M output
47 providers
Moonshot AI: Kimi K2 Instruct 0905 T
Tools JSON Open weights

September 2025 Kimi K2 instruction update with a 256K context and improved agentic coding

moonshotai/kimi-k2-0905 2025-09-05 262.144K context $0.6/M input $2.5/M output
8 providers
Moonshot AI: Kimi K2 Instruct T
Tools JSON Open weights

Non-thinking Kimi K2 MoE instruction model for coding and tool-using agents

moonshotai/kimi-k2-instruct 2025-07-11 131.072K context $0.4/M input $1.8/M output
5 providers