Brutal AI

Brutal AIModelsGuidesUse an uncensored model in OpenCode

Setup4 min read

Use an uncensored model in OpenCode

Declare XField once as a provider in opencode.json with the AI SDK's OpenAI-compatible client, list the models with their limits, and pick one with /models.

What you need

  • OpenCode installed.
  • An XField API key in the environment variable XFIELD_API_KEY. No key yet? Request access.
  • The base URL https://xfield.cloud.brutal.ai/api/v1. It already ends in /v1.

Setup

Use ~/.config/opencode/opencode.json for every project, or opencode.json in a project root for just that one:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "field": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "XField",
      "options": {
        "baseURL": "https://xfield.cloud.brutal.ai/api/v1",
        "apiKey": "{env:XFIELD_API_KEY}"
      },
      "models": {
        "deepseek-v4.1-flash-1017-uncensored": {
          "name": "DeepSeek V4.1 Flash",
          "limit": { "context": 1000000, "output": 65536 }
        },
        "glm-5.3-uncensored": {
          "name": "GLM 5.3",
          "limit": { "context": 1000000, "output": 131072 }
        },
        "glm-5.3-flash-uncensored": {
          "name": "GLM 5.3 Flash",
          "limit": { "context": 1000000, "output": 32768 }
        }
      }
    }
  },
  "model": "field/deepseek-v4.1-flash-1017-uncensored",
  "small_model": "field/glm-5.3-flash-uncensored"
}

Start opencode, run /models and pick from the XField section. small_model handles light work such as session titles, so it should be the cheap one.

Which model to pick

JobModel idWhy
Most coding and agent workdeepseek-v4.1-flash-1017-uncensoredLeads our models on coding and security benchmarks, fast, cheap once the prompt is cached
Hard agentic coding, very long outputskimi-k3.1-pro-uncensoredMoonshot AI's flagship, up to 922K output tokens; costs more, so use it where it matters
Long reports, a second opinionglm-5.3-uncensored128K output, a different model family
Cheap background work, summariesglm-5.3-flash-uncensoredThe lighter GLM, built for volume
Quick chat and draftsqwen-3.8-27b-uncensoredSmallest and fastest to answer

Context and output limits

Tell OpenCode the limits, otherwise it guesses low and condenses too early. Turn image input on only for models that read images.

Model idContextMax outputReads images
deepseek-v4.1-flash-1017-uncensored1M64KYes
kimi-k3.1-pro-uncensored1M922KYes (and video)
glm-5.3-uncensored1M128KNo
glm-5.3-flash-uncensored1M32KYes
qwen-3.8-27b-uncensored128K32KYes

Check the key and URL first

If this returns a JSON list of models, the base URL and key are right and any remaining problem is in the tool's settings:

curl -sS "https://xfield.cloud.brutal.ai/api/v1/models" -H "Authorization: Bearer $XFIELD_API_KEY"

Troubleshooting

  • 401 Unauthorized. XFIELD_API_KEY was not set when OpenCode started; {env:...} then resolves to an empty string.
  • Model not found. The id must match exactly, including -uncensored. List them with the call above.
  • The tool condenses or cuts off too early. It does not know these models. Set context and output from the limits table.

Why an uncensored model in a coding tool

Stock models stop mid-task on legitimate work: an exploit proof of concept for your own bug bounty, a fuzzing harness, malware triage, a detection rule that has to describe the attack it detects. In an agent loop one refusal breaks the whole run. Our models are abliterated: the direction in the weights that leads to refusals is removed, the rest of the model stays the same. How abliteration works.

More setups

Get a key

XField is the model API of Brutal AI in Vienna. It is hosted in the EU, does not store prompts or completions (data handling), and access is reviewed company by company under our acceptable use policy. Request a key.

Access

Access is by invitation and reviewed per company. Tell Lukas what you are building and where your current model refuses. Pricing comes with the invitation.

lukas@brutal.ai

Uncensored models don't change what is allowed: you stay responsible for use and results under our acceptable use policy.

More guides