Kimi K3.1 Pro, run in Germany by Verda.
Kimi K3.1 Pro is an open-weight model by Moonshot AI that reads text and images, released under the Kimi K3.1-Pro License. On Evanesca, Verda runs it in Germany, on its own GPUs, and Evanesca keeps no copy of your prompts.
Checked on 16 Nov 2026
Kimi K3.1 Pro
by Moonshot AI · Kimi K3.1-Pro License
- Input / M
- €3.024
- Output / M
- €15.12
- Quick question
- €0.0070
Prices per million tokens, VAT included. Quick question: 300 tokens in, 400 out.
The model at a glance
- Maker
Moonshot AI
- Licence
- Kimi K3.1-Pro License
- Size
- 2.8T (104B active) parameters
- Context
- 1M tokens as the operator announces it
- Type
- Text + images
- Tool calling
- Yes measured by a real call
- Reasoning
- Yes — it thinks before answering those tokens are billed as output
- Run by
- Verda Verda Cloud Oy, Helsinki — a Finnish company (formerly DataCrunch), backed mainly by Nordic investors
- Where it runs
- Germany
- Checks
- 6/6 All six: Licence, Provenance, Security, GDPR, Benchmarks and Live test
- Price
- €3.024 per million input tokens · €15.12 per million output tokens VAT included; reasoning tokens count as output
- Open weights
- Yes published by the maker: an operator can run the model
- Model card
- huggingface.co/moonshotai/Kimi-K3.1-Pro
Where your prompts go
Here is the route of a message sent to Kimi K3.1 Pro, end to end. No call to the model leaves Europe: not as a backup, not when servers are busy.
- From your browser to EvanescaEvanesca passes your message straight on to Verda. It stores neither the prompt nor the answer: both stay in memory only while the answer is written.
- Verda, GermanyVerda runs Kimi K3.1 Pro in Germany, on its own GPUs and with no sub-processor (Verda, read on 15 Nov 2026). Verda Cloud Oy is a Finnish company in Helsinki: under EU law, it is out of the CLOUD Act's direct reach; its full shareholding is not published. Verda states that it stores no request: context is held briefly in an in-memory inference cache and removed within 2 hours, never used for training and never sent to the model maker or any third party (Verda Inference FAQ).
- Back to you, with its passportThe answer comes back with its passport: model, operator, country, and what was kept. Your history lives in your browser; about this message, Evanesca keeps only billing data — tokens and cost.
What it costs
€3.024 per million input tokens and €15.12 per million output tokens, VAT included. Three typical exchanges on Kimi K3.1 Pro:
| Example | Tokens in | Tokens out | Cost |
|---|---|---|---|
| A quick question | 300 | 400 | €0.0070 |
| A long document to summarise | 15,000 | 800 | €0.0575 |
| A message late in a long conversation | 40,000 | 2,000 | €0.1512 |
€5 of credit covers about 710 quick questions on this model. Reasoning tokens count as output, so a model that thinks first can cost more than these examples. The exact cost of each answer shows under it. All prices.
Licence
Moonshot AI publishes Kimi K3.1 Pro under the Kimi K3.1-Pro License. Read its terms on the official model card.
The official model card is the reference.
Similar models
Same maker first, then the same type.
DeepSeek V4.1 Flash
DeepSeek · Text + images · checks 6/6
€1.008 · €3.024 / M tokens
Mistral Large 4
Mistral AI · Text + images · checks 6/6
€2.5133 · €7.7246 / M tokens
Qwen3.8 27B
Qwen · Text + images · checks 6/6
€0.252 · €1.092 / M tokens
Apertus v1.5 70B
Swiss AI Initiative · Text + images · checks 5/6
€1.2936 · €4.62 / M tokens
Qwen3.6 35B
Qwen · Text + images · checks 6/6
€0.42 · €2.52 / M tokens
Gemma 4 26B
Google · Text + images · checks 6/6
€0.168 · €0.84 / M tokens
Frequently asked questions
Where does Kimi K3.1 Pro run on Evanesca?
In Germany, on Verda's own GPUs, with no sub-processor (Verda, read on 15 Nov 2026). Verda Cloud Oy is a Finnish company in Helsinki, so your prompts stay out of the CLOUD Act's direct reach. Its full shareholding is not published. No call to the model leaves Europe: not as a backup, not when servers are busy.
Does Evanesca keep what I send to Kimi K3.1 Pro?
No. The prompt and the answer stay in memory only while the answer is written, and your history lives in your browser; about your messages, Evanesca keeps only billing data — tokens and cost. Verda states that it stores no request: context is held briefly in an in-memory inference cache and removed within 2 hours, never used for training and never sent to the model maker or any third party (Verda Inference FAQ).
How much does Kimi K3.1 Pro cost?
€3.024 per million input tokens and €15.12 per million output tokens, VAT included. A quick question — 300 tokens in, 400 out — costs about €0.0070. A subscription is optional: credit starts at €5.
Can Kimi K3.1 Pro call tools?
Yes: a real call through Verda returned proper tool calls (function calling).
Which licence is Kimi K3.1 Pro under?
Kimi K3.1-Pro License. Read its terms on the official model card.
Which checks has Kimi K3.1 Pro passed?
6 of 6: Licence, Provenance, Security, GDPR, Benchmarks and Live test. Its passport shows the same stamps; how each check works.
Does Moonshot AI see what I type?
No. Moonshot AI published the weights, and Verda runs them in Germany: your prompts go to Verda's servers, not to Moonshot AI. Verda adds that “no request data is sent to the model developer or any other third party” (Verda's FAQ, read on 15 Nov 2026).