Hermes Agent with a custom OpenAI-compatible endpoint: connect Heabsy
Hermes Agent accepts any OpenAI-compatible endpoint as a custom provider. Run hermes model, choose "Custom endpoint (self-hosted / VLLM / etc.)" and enter https://api.heabsy.com/v1, your key and a model ID such as qwen38 — or write the same into ~/.hermes/config.yaml. Hermes needs tool calling and at least 64,000 tokens of context; every model in the table has both.
model:
default: qwen38
provider: custom
base_url: https://api.heabsy.com/v1
key_env: HEABSY_API_KEY
context_length: 262144https://api.heabsy.com/v1 · Anthropic base URL https://api.heabsy.comInteractive setup
- Run
hermes model. - Choose "Custom endpoint (self-hosted / VLLM / etc.)".
- Base URL
https://api.heabsy.com/v1, your Heabsy key, model IDqwen38. - Hermes writes the result to
~/.hermes/config.yaml, the single source of truth for model and endpoint.
Key and context window
key_env: HEABSY_API_KEY reads the key from the environment, so it never lands in the file; api_key: works too. Hermes rejects windows under 64,000 tokens at startup and cannot always detect the window of a custom endpoint — set context_length to the model's window from the table.
Which model should I pick?
Hermes runs multi-step loops with tools, so every model needs tool calling. qwen38 is a good default; glm53 gives a 1,048,576-token window for long sessions at $0.52 / $0.96 per million input / output tokens. Both run on our own GPUs in the EEA and store no prompts.
| Model | ID | Context | Max output | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| Qwen3.8 27B Fable | qwen38-fable | 512K | 32K | $0.095 | $0.58 |
| Qwen3.8 27B Queen | qwen38-queen | 512K | 32K | $0.095 | $0.58 |
| GLM 5.3 Flash Uncensored | glm53-flash-unc | 1M | 128K | $0.1739 | $0.6957 |
| MiMo V2.6 Flash Uncensored | mimo-v26-flash-unc | 1M | 64K | $0.4404 | $1.4285 |
| MiMo V2.6 Flash | mimo-v26-flash | 1M | 64K | $0.4762 | $1.4286 |
| MiMo V2.6 Flash Abliterated | mimo-v26-flash-abl | 1M | 64K | $0.4762 | $1.4286 |
| GLM 5.3 | glm53 | 1M | 128K | $0.5217 | $0.9565 |
| Qwen3.8 27B Uncensored Fast by Heabsy | qwen38-unc-fast | 512K | 230K | $0.7904 | $2.4952 |
| Qwen3.8 27B Uncensored | qwen38-unc | 1M | 230K | $0.8095 | $2.5238 |
| Qwen3.8 27B | qwen38 | 256K | 32K | $1.00 | $3.00 |
| GLM 5.3 Uncensored | glm-5.3-uncensored | 1M | 128K | $1.1905 | $2.1429 |
| CyberHeabsy | cyberheabsy | 1M | 64K | $1.4286 | $4.2857 |
All models above run on Heabsy hardware in the EEA; prompts and completions are not stored. Thecatalog also lists 23 partner-hosted models — Claude, GPT-6, Kimi K3, DeepSeek V4 and others — with the same key and invoice, but without zero data retention and without an EEA guarantee.
Verify it works
Start Hermes and ask it to list the files in the current folder. The tool call should execute and return the list — if it appears as plain text instead, the model is not calling tools.
If the tool shows an error you cannot read, test the key on its own first — one short request outside the tool:
curl https://api.heabsy.com/v1/chat/completions \
-H "Authorization: Bearer $HEABSY_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"qwen38","max_tokens":16,"messages":[{"role":"user","content":"Say hi"}]}'Troubleshooting
| What you see | What to change |
|---|---|
| Tool calls appear as text instead of executing | Pick a model with tool calling from the table; all of ours have it. |
| Hermes refuses to start because the context is too small | Set context_length in config.yaml to the model's window. |
| 404 Not Found | base_url must end in /v1 — no /chat/completions. |
| 401 Unauthorized | The key in the variable named by key_env is wrong or revoked — copy it again from the console. |
| 403 with an HTML page ("Sorry, you have been blocked") | No key was sent at all: the key variable is empty or not set where the tool runs. Requests without a key are stopped before they reach the API; a wrong key gets a JSON 401 instead. |
Where do my prompts go?
From Hermes Agent straight to api.heabsy.com. On our own models the request is processed in the European Economic Area and the text is not stored — we keep what billing needs: the model, the token counts and the time. Partner-hosted models are processed by third-party inference providers; each is marked in the catalog, and none of them carries the no-retention promise. The terms are in the data processing agreement, the invoice comes from an EU company.
FAQ
Can I use Hermes Agent with Qwen 3.8 27B?+
Yes: qwen38, 262,144 tokens of context, tool calling, $1 / $3 per million input / output tokens, served on our own GPUs in the EEA. Use the config block above as it is.
Can I use partner models such as Kimi K3 in Hermes?+
Yes, with the same key — kimi-k3 has tool calling and a 1,048,576-token window. It is partner-hosted: no zero data retention and no EEA guarantee.
Where do my prompts go?+
From Hermes on your machine to api.heabsy.com. Our own models process them in the EEA and do not store the text; partner-hosted models follow their providers' policies.
Also set up:Claude Code · OpenCode · Cline
Hermes Agent is developed by Nous Research; Heabsy is not affiliated with Nous Research.