> ## Content Index
> Fetch the complete content index at: https://www.tao.media/llms.txt
> Use this file to discover other available public pages before exploring further.

# Actual Computer Added as Native Inference Provider for Hermes Agent
- URL: https://www.tao.media/actual-computer-added-as-native-inference-provider-for-hermes-agent/
- Published: 2026-08-06T20:14:51.000Z
- Updated: 2026-08-06T20:27:57.000Z
- Description: Hermes Agent users can now connect Actual Computer as an inference provider, extending the Bittensor subnet’s private compute model into Nous Research’s open-source agent stack.
- Author: Bart Hillerich
- Tags: Actual Computer, News, Subnets, Nous Research, AI

This is (another) big one for Bittensor.

[Actual Computer](https://actual.inc/?ref=tao.media) has been added as a first-class native inference provider for Hermes Agent, giving users a new way to route AI model requests through their own hardware.

The integration lets Hermes users [configure Actual](https://hermes-agent.nousresearch.com/docs/integrations/providers?ref=tao.media) either through Actual's hosted relay or through a local daemon. A Hermes user can then run inference from a personal or private compute setup instead of relying only on a standard cloud model API, and they can access their personal inference capacity from anywhere.

The [announcement post](https://x.com/actualinc/status/2085172429895172136?s=20&ref=tao.media) from Actual has gone viral, racking up over 320k impressions on X. That's big visibility for Bittensor-native tech, only about 1.5 weeks after Engy's viral Kimi K3 viral update. Subnets are heating up in the broader AI conversation, and we think there's much more to come.

[Is Engy the Breakout Subnet Bittensor Has Been Waiting For?After Kimi released the K3 model weights, Engy founder Ning said the team served the 2.8T-parameter model on 80 RTX 5090s, offering a striking proof point for frontier open-model inference on consumer hardware.![](https://storage.ghost.io/c/78/0b/780ba906-b1a7-4bf0-873c-bdd5c32e5331/content/images/icon/Group-1321319358-cb868f20-7926-4037-92da-b884e6f871be.png)Intelligence | Bittensor News, Insights, StoriesBart Hillerich![](https://storage.ghost.io/c/78/0b/780ba906-b1a7-4bf0-873c-bdd5c32e5331/content/images/thumbnail/27cb6a11-2c7c-4697-815e-bd310688de87-af922747-a585-4135-8f37-88a2854e664c.png)](https://www.tao.media/is-engy-the-breakout-subnet-bittensor-has-been-waiting-for/)

Tom Lynch, Actual Computer's owner, welcomed the release, writing that he uses Hermes frequently and that the connection had been "a huge goal" for him in [a post on X](https://x.com/Tom%5FA%5FLynch/status/2085173818180710683?s=20&ref=tao.media).

> "Inference no longer means raw completions. Now chat is waining and soon everyone (including the model) will expect their inference to happen inside of a harness. I use hermes agent every day and im thrilled to have it work out of the box with actual now."

Hermes Agent ranks as the most-used global AI agent by token consumption on [OpenRouter](https://openrouter.ai/apps/hermes-agent?ref=tao.media), processing over a trillion daily tokens.