Skip to main content

shift-cli inference

Inference — AI model inference service.

Usage​

shift-cli inference <subcommand> [arguments] [flags]

Description​

The Inference CLI provides access to the platform's AI inference capabilities. It interfaces with the Inference API mounted at /api/v1/inference/* on the gateway.

Subcommands​

Provider Management​

# List registered inference providers
shift-cli inference providers list [--json]

# Add a provider
shift-cli inference providers add \
  --name openai-prod \
  --type openai \
  --default-model gpt-4o-mini \
  [--base-url https://api.openai.com/v1] \
  [--json]

# Remove a provider
shift-cli inference providers rm <id-or-name> [--json]

Model Sync​

# Sync available models from a provider
shift-cli inference models sync <provider-id-or-name> [--json]

# List synced models
shift-cli inference models list [--provider <id>] [--json]

Chat​

# Send a chat completion request
shift-cli inference chat \
  --model gpt-4o-mini \
  --message "Summarize this project" \
  [--provider openai-prod] \
  [--json]

Embeddings​

# Generate text embeddings
shift-cli inference embed \
  --model text-embedding-3-small \
  --input "Quarterly revenue forecast" \
  [--json]

Images​

# Generate an image
shift-cli inference image \
  --model gpt-image-1 \
  --prompt "Architecture diagram" \
  [--json]

Transcription​

# Submit a transcription job
shift-cli inference transcribe \
  --model whisper-1 \
  --source https://example.com/recording.mp3 \
  [--language en] \
  [--json]

Usage Reporting​

# View usage statistics
shift-cli inference usage [--provider openai-prod] [--since 2026-03-01] [--limit 25] [--json]

Examples​

# Register OpenAI and sync models
shift-cli inference providers add --name openai --type openai --json
shift-cli inference models sync openai --json

# Quick chat completion
shift-cli inference chat --model gpt-4o-mini --message "Hello" --json

# Check monthly usage
shift-cli inference usage --since 2026-03-01 --json