The Orchestrator › Models › Moonshot AI
Moonshot AI: MoonshotAI: Kimi K2.5
Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...
Specifications
| Provider | Moonshot AI |
| Model ID | moonshotai/kimi-k2.5 |
| Context window | 262K tokens (262,144) |
| Modalities | text + image → text |
| Tokenizer | Other |
| Router sources | OpenRouter, TrustedRouter, Vercel AI Gateway |
| Released | 2026-01-27 |
| Tracked since | 2026-11-14 |
| Popularity | Not in the ranked set |
| Reasoning | On by default |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | moonshotai/kimi-k2.5 | $0.5 | $2.50 | 262K |
| standard | TrustedRouter | moonshotai/kimi-k2-5 | $0.591 | $3.69 | 256K |
| standard | Vercel AI Gateway | moonshotai/kimi-k2.5 | $0.6 | $3.00 | 256K |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.6 | $3.00 | 256K |
| 2026-11-14 | $0.591 | $3.69 | 256K |
| 2026-11-14 | $0.475 | $2.37 | 262K |
| 2026-11-14 | $0.5 | $2.50 | 262K |
Capabilities
Vision (image input)Agentic tool useStructured outputsReasoning mode256K+ context windowControllable reasoningNo content storage (attested)
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p