> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/circuitbreakerlabs/cli/llms.txt
> Use this file to discover all available pages before exploring further.

# Configuration

> Configure the Circuit Breaker Labs CLI with environment variables and global options

The Circuit Breaker Labs CLI can be configured through environment variables, command-line flags, and provider-specific options.

## Environment Variables

### Required Variables

#### CBL\_API\_KEY

Your Circuit Breaker Labs API key for authenticating with the evaluation service.

```bash theme={null}
export CBL_API_KEY="cbl_your_api_key_here"
```

<Note>
  Get your API key by contacting [team@circuitbreakerlabs.ai](mailto:team@circuitbreakerlabs.ai?subject=Getting%20Set%20Up)
</Note>

### Optional Variables

#### CBL\_API\_BASE\_URL

Custom base URL for the Circuit Breaker Labs API. Defaults to `wss://api.circuitbreakerlabs.ai/v1`.

```bash theme={null}
export CBL_API_BASE_URL="wss://custom-endpoint.example.com/v1"
```

#### OPENAI\_API\_KEY

Required when using the OpenAI provider.

```bash theme={null}
export OPENAI_API_KEY="sk-your_openai_key_here"
```

#### OPENAI\_BASE\_URL

Custom base URL for OpenAI-compatible endpoints. Defaults to `https://api.openai.com/v1`.

```bash theme={null}
export OPENAI_BASE_URL="https://api.openai.com/v1"
```

#### OPENAI\_ORG\_ID

Optional OpenAI organization ID.

```bash theme={null}
export OPENAI_ORG_ID="org-your_org_id"
```

#### OLLAMA\_BASE\_URL

Base URL for Ollama server. Defaults to `http://localhost:11434`.

```bash theme={null}
export OLLAMA_BASE_URL="http://localhost:11434"
```

## Global Options

These flags can be used with any evaluation command and must be specified before the evaluation type.

### --log-level

Set the logging verbosity level.

```bash theme={null}
cbl --log-level debug single-turn ...
```

**Values:** `error`, `warn`, `info` (default), `debug`, `trace`

### --log-mode

Enable log mode to disable the TUI and output logs to stdout instead. Useful for CI/CD pipelines and debugging.

```bash theme={null}
cbl --log-mode single-turn ...
```

<Warning>
  In log mode, the interactive TUI is disabled. Progress will be logged as text output instead.
</Warning>

### --output-file

Specify a custom output file path for evaluation results. By default, results are saved to auto-generated files with timestamps:

* Single-turn: `circuit_breaker_labs_single_turn_evaluation_YYYYMMDD_HHMMSS.json`
* Multi-turn: `circuit_breaker_labs_multi_turn_evaluation_YYYYMMDD_HHMMSS.json`

```bash theme={null}
cbl --output-file results/my-evaluation.json single-turn ...
```

### --add-header

Add custom HTTP headers to provider requests. Can be specified multiple times for multiple headers.

```bash theme={null}
cbl --add-header "X-Custom-Header:value1" --add-header "Authorization:Bearer token" single-turn ...
```

**Format:** `"Key:Value"`

## Provider Configuration

### OpenAI Provider

#### Required Options

```bash theme={null}
cbl single-turn openai \
  --api-key "sk-..." \        # Or use OPENAI_API_KEY env var
  --model "gpt-4o"
```

#### Optional Parameters

<Accordion title="Temperature and Sampling">
  **--temperature** `<float>`\
  Sampling temperature between 0 and 2. Higher values make output more creative.

  ```bash theme={null}
  --temperature 0.7
  ```

  **--top-p** `<float>`\
  Nucleus sampling parameter. Alternative to temperature.

  ```bash theme={null}
  --top-p 0.9
  ```

  **--frequency-penalty** `<float>`\
  Number between -2.0 and 2.0. Positive values penalize repeated tokens based on frequency.

  ```bash theme={null}
  --frequency-penalty 0.5
  ```

  **--presence-penalty** `<float>`\
  Number between -2.0 and 2.0. Positive values penalize repeated tokens based on presence.

  ```bash theme={null}
  --presence-penalty 0.5
  ```
</Accordion>

<Accordion title="Token Control">
  **--max-completion-tokens** `<integer>`\
  Maximum number of tokens to generate in the completion.

  ```bash theme={null}
  --max-completion-tokens 1000
  ```

  **--stop** `<string,string,...>`\
  Up to 4 sequences where the API will stop generating tokens.

  ```bash theme={null}
  --stop "END,STOP"
  ```

  **--logit-bias** `<token_id:bias,...>`\
  Modify likelihood of specified tokens appearing. Bias values between -100 and 100.

  ```bash theme={null}
  --logit-bias "50256:-100,13:5"
  ```
</Accordion>

<Accordion title="Advanced Options">
  **--n** `<integer>`\
  Number of chat completion choices to generate for each input.

  ```bash theme={null}
  --n 3
  ```

  **--logprobs** `<bool>`\
  Return log probabilities of output tokens.

  ```bash theme={null}
  --logprobs true
  ```

  **--top-logprobs** `<integer>`\
  Number between 0 and 20 specifying most likely tokens to return.

  ```bash theme={null}
  --top-logprobs 5
  ```

  **--service-tier** `<auto|default|flex|scale|priority>`\
  Specifies the processing tier for the request.

  ```bash theme={null}
  --service-tier scale
  ```

  **--reasoning-effort** `<none|minimal|low|medium|high|xhigh>`\
  Constrains effort on reasoning for reasoning models.

  ```bash theme={null}
  --reasoning-effort high
  ```

  **--store** `<bool>`\
  Whether to store the output of this chat completion request.

  ```bash theme={null}
  --store true
  ```
</Accordion>

### Ollama Provider

#### Required Options

```bash theme={null}
cbl single-turn ollama \
  --model "llama2"
```

#### Optional Parameters

<Accordion title="Model Options">
  **--temperature** `<float>`\
  Model temperature. Higher values = more creative (default: 0.8).

  ```bash theme={null}
  --temperature 0.7
  ```

  **--top-k** `<integer>`\
  Reduces probability of generating nonsense. Higher = more diverse (default: 40).

  ```bash theme={null}
  --top-k 50
  ```

  **--top-p** `<float>`\
  Works with top-k. Higher values = more diverse text (default: 0.9).

  ```bash theme={null}
  --top-p 0.95
  ```

  **--num-predict** `<integer>`\
  Maximum tokens to predict (default: 128, -1 = infinite, -2 = fill context).

  ```bash theme={null}
  --num-predict 256
  ```

  **--num-ctx** `<integer>`\
  Size of the context window (default: 2048).

  ```bash theme={null}
  --num-ctx 4096
  ```
</Accordion>

<Accordion title="Sampling Control">
  **--mirostat** `<0|1|2>`\
  Enable Mirostat sampling (0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0).

  ```bash theme={null}
  --mirostat 2
  ```

  **--mirostat-eta** `<float>`\
  Mirostat learning rate (default: 0.1).

  ```bash theme={null}
  --mirostat-eta 0.15
  ```

  **--mirostat-tau** `<float>`\
  Controls balance between coherence and diversity (default: 5.0).

  ```bash theme={null}
  --mirostat-tau 6.0
  ```

  **--repeat-penalty** `<float>`\
  How strongly to penalize repetitions (default: 1.1).

  ```bash theme={null}
  --repeat-penalty 1.2
  ```

  **--repeat-last-n** `<integer>`\
  How far back to look to prevent repetition (default: 64, 0 = disabled, -1 = num\_ctx).

  ```bash theme={null}
  --repeat-last-n 128
  ```

  **--tfs-z** `<float>`\
  Tail free sampling - reduces impact of less probable tokens (default: 1).

  ```bash theme={null}
  --tfs-z 0.95
  ```
</Accordion>

<Accordion title="Hardware and Performance">
  **--num-gpu** `<integer>`\
  Number of layers to send to GPU(s).

  ```bash theme={null}
  --num-gpu 32
  ```

  **--num-thread** `<integer>`\
  Number of threads to use during computation.

  ```bash theme={null}
  --num-thread 8
  ```

  **--num-gqa** `<integer>`\
  Number of GQA groups in transformer layer.

  ```bash theme={null}
  --num-gqa 8
  ```
</Accordion>

<Accordion title="Other Options">
  **--seed** `<integer>`\
  Random number seed for generation (default: 0).

  ```bash theme={null}
  --seed 42
  ```

  **--stop** `<string>`\
  Stop sequences (can be specified multiple times).

  ```bash theme={null}
  --stop "END"
  ```

  **--logprobs** `<bool>`\
  Return log probabilities for each token.

  ```bash theme={null}
  --logprobs true
  ```
</Accordion>

### Custom Provider

For APIs that aren't OpenAI-compatible, use Rhai scripts to translate between schemas.

```bash theme={null}
cbl single-turn custom \
  --url "https://api.example.com/v1/chat" \
  --script ./my-provider.rhai
```

**--url** `<string>` (required)\
Endpoint URL to POST requests to.

**--script** `<path>` (required)\
Path to the Rhai script file that handles request/response translation.

<Note>
  See the [examples/providers/](https://github.com/circuitbreakerlabs/cli/tree/main/examples/providers) directory for example Rhai scripts.
</Note>

## Complete Example

```bash theme={null}
export CBL_API_KEY="cbl_your_key"
export OPENAI_API_KEY="sk-your_key"

cbl \
  --log-level info \
  --output-file evaluations/test-run.json \
  --add-header "X-Request-ID:test-123" \
  single-turn \
  --threshold 0.5 \
  --variations 3 \
  --maximum-iteration-layers 2 \
  --test-case-groups suicidal_ideation \
  openai \
  --model gpt-4o \
  --temperature 0.7 \
  --max-completion-tokens 500
```