Use your Venice plan credits for AI Butler

Use your Venice plan credits for AI Butler

Updated October 17, 2026 · Meowport Pro · About 7 minutes

Venice is a privacy-focused AI service that says it never stores your prompts or responses. Its paid plans include credits that can be spent on its API, and that API speaks the OpenAI chat-completions format, so Meowport's AI Butler can use it with one key, one address and one model name. This guide shows the setup, what the credits cover, and two things that make Venice a less certain fit than other providers.

Who this is for

• You have Meowport Pro. AI Butler is a Pro feature; a custom model service only chooses which model answers.
• You pay for Venice Pro, Pro Plus or Max and want the credits that come with it to cover the Butler.
• You value Venice's privacy stance and accept that your requests still leave your device and go to Venice and, for some models, to a further provider.

What the credits cover

• Venice meters everything in credits: 100 credits equal US$1. Pro (US$18 a month) includes 100 credits a month for premium models, media and the API, which is one US dollar of API use, plus a one-time welcome balance of a few dollars that Venice's pages quote differently. Pro Plus (US$68) includes US$75 of monthly API spend with two months of banking, and Max (US$200) includes US$225 with three months of banking.
• Venice is explicit that paid plans do not include unlimited API use: API calls are billed per token, and plan credits refresh monthly.
• Beyond the plan credits you can fund a US-dollar balance; those purchased credits never expire. Venice also offers a daily allowance to people who stake its token; that is a separate mechanism described in Venice's docs and not needed here.
• Spending order is the staking allowance first, then plan credits, then your dollar balance. When nothing is left, the API answers with a 402 "insufficient balance" error and the Butler's request fails.

Step 1: Create a Venice API key

1. Sign in to Venice and open https://venice.ai/settings/api.
2. Choose Generate New API Key. Pick the Inference Only type; give it a name such as "Meowport" and, if you like, an expiration date.
3. Optionally set an epoch consumption limit, which caps what the key may spend per day.
4. Tap Generate and copy the key. Venice shows the full key once.

To revoke a key later, delete it on the same page.

Step 2: Connect it in Meowport

1. Open Meowport and go to Settings → AI Butler.
2. Choose Custom model API (marked Advanced).
3. Fill in the three fields:
   • Base URL: https://api.venice.ai/api/v1
   • API Key: paste the key from Step 1.
   • Model name: a model that reads images is the most useful for the Butler, for example mistral-small-3-2-24b-instruct (inexpensive) or qwen3-vl-235b-a22b (larger). Text-only models such as llama-3.3-70b work for typing but cannot read reports.
4. Tap Fetch models to confirm the key works, and pick a model from the list if you prefer.
5. Tap Save model settings.

How Meowport treats this provider

Meowport recognizes a handful of providers by their address and gives those models its full trust. Venice is not one of them yet, so the Butler runs it in its cautious mode: it checks the model's answer against its own rules before acting, it does not let the model look up extra records on its own, and it may ask you more often or decline an unusual request.

Two Venice specifics to know:

• Venice adds its own system prompt to every API request unless an application turns it off, and Meowport does not send Venice-specific settings. That extra prompt may make the Butler's structured replies less reliable with some models. If the Butler often reports a malformed reply, try a larger model.
• Venice documents a newer name for the reply-length setting and may reject unknown fields. Meowport uses the common setting that Venice still accepts at the time of writing; if that changes, the Butler will show Venice's error.

To read a clinic report or a lab photo, tap Test this model's image support once in the settings. Venice rejects images on models that cannot see; Meowport verifies the model you chose before it offers the attachment.

Step 3: Try it

Open the Butler and say something small, for example "Mochi weighed 4.2 kg this morning." The Butler drafts a weight record and, depending on your Before saving setting, saves it or asks first. Whatever you choose, it always asks before changing a record you already have, and it never deletes one. Anything it saves can be undone from the chat.

Costs and limits

• Venice publishes per-token prices in US dollars. At the time of writing a small model such as mistral-small-3-2-24b-instruct costs a fraction of a cent per Butler request; larger models cost more. The pricing page lists every model.
• Rate limits depend on the model's size class, from a few hundred requests per minute downwards; all are far above what the Butler sends.
• Plan credits refresh monthly; purchased dollar credits do not expire. When both are gone the API returns 402 and Meowport shows the message. Nothing is written.

Privacy and what is sent

With a custom model service, each request sends your message, any images you attached, and the pet records the Butler needs for that request to the provider, under the provider's terms and privacy policy. Venice states that it does not store or log prompt and response content for normal inference, and it labels models as "private", where the content is not retained after the request, or "anonymized", where the model's provider still sees the prompt. Request metadata such as key ids and token counts is kept separately. Meowport's privacy policy describes the custom-provider path. Apple's on-device model stays available: switch back any time in the same settings. To stop using Venice, clear the key in Meowport and delete it in Venice's settings.

Troubleshooting

• Authentication error. The key was pasted incompletely, has expired or was deleted. Generate a new one.
• 402 or "insufficient balance". Your plan credits and dollar balance are used up, or the key's daily consumption limit was reached. Check the balance in Venice.
• Images are rejected. The model cannot see. Run Test this model's image support or choose a vision model.
• Replies are often malformed. Venice's added system prompt or a small model is the likely cause. Switch to a larger model, or to a provider Meowport recognizes.

Your responsibilities

• Check the provider's pricing and terms yourself; this guide reflects October 17, 2026 and cannot follow later changes.
• Keep the API key private. Do not paste it into chats, screenshots or shared notes; set an expiration or a daily limit if you can.
• Watch your balance in Venice. You, not Meowport, pay for what the key spends.
• What you send is handled by Venice and, for anonymized models, by a further provider under their own policies, outside Meowport's control.
• AI can be wrong. Review what the Butler writes, and talk to your veterinarian before any medical decision.
• A Meowport is a personal record, not a certificate.