HammerAI proxy mode lets a character chat run on a model HammerAI does not host. This guide fills its four Proxy fields with an uncensored roleplay build: the URL, the model id, the context length and the key. Nothing to download and no GPU on your side. The field names come from HammerAI's Proxy Mode doc, read on 24 October 2026.
If builds with the refusals edited out are new to you, start with AI with no restrictions, then come back here to connect one.
What HammerAI is
HammerAI is a character chat and story app on the web and on desktop for macOS, Windows and Linux. Its own docs describe it as a platform for roleplay, image generation and interactive stories run on language models. The desktop app runs models on your own machine, and HammerAI also hosts cloud models, more of them on its paid plans.
Proxy is the third option. It sends every prompt to an OpenAI-compatible API you name, so the model, its context window and its price are the API's, not HammerAI's. HammerAI's pricing page lists "Proxy to any LLM API" on the Free plan and on every paid plan, as read on 24 October 2026. The same page caps HammerAI's own cloud models at 8,192 tokens of context on Free and 65,536 on its top plan; a proxied model is bounded by its own window instead.
The four HammerAI Proxy fields
Selecting Proxy reveals four fields. These are the values for this API; the labels are the ones in HammerAI's Proxy Mode doc, read on 24 October 2026.
- Proxy URL
- https://api.uncensoredllm.ai/v1The base URL, ending in /v1
- Proxy Model
- cydonia-24b-v4.3Any id from the Models page, typed exactly
- Proxy Context Length
- 131072Optional; the window from the table below
- Proxy API Key
- sk-ullm-live-…Created in the console, shown once
The doc's own examples use the same shape for other services: a base URL that ends in /v1, never one that ends in /chat/completions. Proxy Model has no dropdown, so the id has to match the catalog character for character.
Connect HammerAI to the uncensored API
- Create a key in the console. It starts with
sk-ullm-and is shown once, so paste it somewhere safe first. The balance is prepaid from $10; a key with nothing on it is accepted but the first message answers 402. - Pick a model id from the uncensored model list or from the table in the next section. Copy it rather than typing it.
- In HammerAI, open any character chat and go to the Settings tab in the right-hand panel.
- Open the LLM dropdown, scroll to the bottom of the list and select Proxy. The four fields appear.
- Paste
https://api.uncensoredllm.ai/v1into Proxy URL. - Paste the id into Proxy Model, for example
cydonia-24b-v4.3. - Enter the model's window in Proxy Context Length, or leave it blank when the catalog publishes none. HammerAI's doc says to leave it blank if unknown.
- Paste the key into Proxy API Key.
- Send a message. If nothing comes back, the doc's advice is to check the endpoint, the model name and the key, and notes that a full reload of the app may be needed for some services to connect.
From then on HammerAI routes every prompt and completion in that chat through the Proxy instead of its own models. Nothing is downloaded and no GPU is involved on your side: the model runs on the API.
Which uncensored build to pick for HammerAI
All four builds below carry the roleplay tag in the catalog, and all four take the OpenAI wire format that HammerAI's Proxy speaks. The prices and windows are read from the live catalog when the page renders.
| Wire formats | |||||
|---|---|---|---|---|---|
| Cydonia 24B v4.3 | Roleplay | $0.484 | $0.60 | 131K | |
| Magidonia 24B v4.3 | Roleplay | $0.3535 | $0.4235 | 33K | |
| Gemma 4 31B Fabled | Roleplay | $0.35 | $1.575 | 262K |
| Model id | Context window in the catalog | Proxy Context Length to enter |
|---|---|---|
cydonia-24b-v4.3 | 131,072 tokens | 131072 |
magidonia-24b-v4.3 | 32,768 tokens | 32768 |
gemma-4-31b-fabled | 262,144 tokens | 262144 |
gemma-4-31b-fabled is one of the Gemma 4 tunes; the rest of that family is on the uncensored Gemma page. A window is a ceiling, not a target: HammerAI sends as much of the chat as the context length allows, and every token it sends is billed as input on each message. Setting the field lower than the window keeps long chats cheaper.
What an evening of HammerAI chat costs
The worked example: 200 messages in one sitting, each reply about 250 tokens. The short column assumes about 4,000 tokens of character and history per message; the long column assumes 16,000, a lorebook and a long chat.
| Model | 200 messages at 4,000 context | 200 messages at 16,000 context |
|---|---|---|
| Cydonia 24B v4.3 | $0.42 | $1.58 |
| Magidonia 24B v4.3 | $0.30 | $1.15 |
| Gemma 4 31B Fabled | $0.36 | $1.20 |
Input dominates: 200 × 4,000 is 800,000 input tokens against 50,000 output tokens. A shorter character description and a lower Proxy Context Length move the bill more than the choice of model. Every per-token rate is on the pricing page.
Check the key and the model id before HammerAI
Before you open HammerAI, this request lists the model ids your key can reach. It costs nothing and takes a second. Swap in your key and run it from a terminal.
curl https://api.uncensoredllm.ai/v1/models \ -H "Authorization: Bearer sk-ullm-live-…"A 200 with a list of ids means the key is good. If the id you put in Proxy Model is in that list, the only field left to get wrong is the URL. A 401 means the key, and the message says which way:
HTTP 401{"error":{"message":"No API key was sent.","type":"authentication_error","param":null,"code":"invalid_api_key"}}HTTP 401{"error":{"message":"The API key is not valid.","type":"authentication_error","param":null,"code":"invalid_api_key"}}Troubleshooting HammerAI Proxy errors
401: the key
"No API key was sent." means Proxy API Key is empty. "The API key is not valid." means it is wrong, cut short or revoked. Create a fresh key in the console and paste it whole.
404 model_not_found: the model id
The id in Proxy Model is not in the catalog: a typo, a display name instead of an id, or a build that has since left. Run the request above and copy the id from its output.
402: the balance
The prepaid balance is at zero. Top up in the console; a request refused with 402 is not charged.
Browser errors (CORS)
On the web app the browser calls the API directly. The API allows calls from any origin, including the Authorization header (checked on 24 October 2026), so there is nothing to switch on. HammerAI's doc mentions a CORS setting only for a local LM Studio server, which does not apply here.
The daily real-model check sends a fixed prompt to every model each day, so the model behind an id is the one on its label. Base URL, key and model id for other clients are on the docs page.
Questions
HammerAI’s pricing page lists "Proxy to any LLM API" on the Free plan and on every paid plan (read 24 October 2026). The model behind the Proxy is billed by the API you connect: here per token, prepaid from $10. See per-token prices.
No. With Proxy selected, HammerAI sends prompts to the API and the model runs there. A GPU only matters for HammerAI’s local models in the desktop app.
The base URL ending in /v1: https://api.uncensoredllm.ai/v1. Do not add /chat/completions; HammerAI’s own examples are base URLs.
The model’s context window from the catalog, or a lower number to keep long chats cheaper. Leave it blank when the catalog publishes no window, as HammerAI’s doc suggests.
Any id that takes the OpenAI wire format works. For character chat, start with a build the catalog tags roleplay; the table above lists four with their windows and prices. See the uncensored model list.
Yes. The builds here are abliterated or uncensored fine-tunes, so the change is in the weights, not in a prompt, and there is no jailbreak text to paste into the character. See AI with no restrictions, explained.