
Four new AI models on xKiro: DeepSeek V4 Flash Vision, Hy3, Muse Spark 1.2 and OX Alpha (free)
DeepSeek V4 Flash Vision · Hy3 · Muse Spark 1.2 · OX Alpha (free)
New Flash model with a 1M-token context, image understanding and tool calling. Free tier for everyday work, PAYG when you need it in production.
No end date announced

DeepSeek V4.1 Flash is available on xKiro through two access tiers: a free tier for everyday work and a pay-as-you-go tier for production traffic. It is the first DeepSeek model built on the company's Causal Encoder-Decoder (CED) architecture, a sparse mixture-of-experts design. On xKiro it keeps the full 1,048,576-token context window, accepts image input alongside text, supports tool and function calling, and lets you switch reasoning on or off per request. Both tiers use the same model and the same context window; the difference is how you pay for it and how much throughput you get. No card is needed to try the free tier.
Current list prices per 1M tokens, read live from the model catalog.
| Model | Context | Input | Output |
|---|---|---|---|
DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flash · 1.05M | 1.05M | $0.075 | $0.3 |
deepseek/deepseek-v4.1-flash:freeNot in the current catalog | — | — | — |
How to use it
The offer applies on every call — no coupon code, no plan change. One endpoint and one key for every model, through any OpenAI- or Anthropic-compatible SDK.
Quickstart guidecurl https://api.xkiro.com/v1/chat/completions \ -H "Authorization: Bearer $XKIRO_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek/deepseek-v4.1-flash", "messages": [ { "role": "user", "content": "Explain HTTP caching in two sentences." } ] }'Keep exploring
Every offer applies on every request through the same OpenAI- and Anthropic-compatible API. 45 models are free with an account.
No credit card · Cancel anytime · 45 free models, free forever