Aider

Pass the base URL and API key as flags. Note: the model must use the openai/ prefix when pointing at a custom endpoint:

aider --openai-api-base https://imbutus.com/v1 \
      --openai-api-key <your-api-key> \
      --model openai/huihui-ai/Huihui-Qwen3.5-9B-Claude-4.6-Opus-abliterated

Or set environment variables permanently, then run aider with the model flag:

export OPENAI_API_BASE=https://imbutus.com/v1
export OPENAI_API_KEY=<your-api-key>
aider --model openai/huihui-ai/Huihui-Qwen3.5-9B-Claude-4.6-Opus-abliterated

Available model IDs (use after openai/ prefix):

openai/huihui-ai/Huihui-Qwen3.5-9B-Claude-4.6-Opus-abliterated
openai/huihui-ai/Huihui-Qwythos-9B-Claude-Mythos-5-1M-abliterated
openai/huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated
openai/huihui-ai/Huihui-Qwen3-Coder-Next-abliterated

Create .aider.model.metadata.json in your project root so Aider knows each model's context window:

{
  "openai/huihui-ai/Huihui-Qwen3.5-9B-Claude-4.6-Opus-abliterated": {
    "max_input_tokens": 65536
  },
  "openai/huihui-ai/Huihui-Qwythos-9B-Claude-Mythos-5-1M-abliterated": {
    "max_input_tokens": 65536
  },
  "openai/huihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated": {
    "max_input_tokens": 131072
  },
  "openai/huihui-ai/Huihui-Qwen3-Coder-Next-abliterated": {
    "max_input_tokens": 262144
  }
}