Deployments API

{
  "model": "Qwen/Qwen2.5-0.5B-Instruct",
  "source": "huggingface",
  "name": "my-deploy",
  "gpu": {"gpu_type": "L4-24GB", "gpu_count": 1},
  "provider": "aws"
}

{
  "error": {
    "code": "invalid_self_heal_attempts",
    "message": "self_heal_attempts must be between 0 and 5 (got 9); 0 turns self-healing off"
  }
}

{
  "object": "deployment",
  "id": "3f9c2a7d1b04",
  "status": "queued",
  "name": "my-deploy",
  "model": "Qwen/Qwen2.5-0.5B-Instruct",
  "source": "huggingface",
  "endpoint_url": null,
  "gpu": {"type": "L4-24GB", "count": 1},
  "provider": "aws",
  "total_requests": 0,
  "error": null,
  "failure_code": null,
  "self_heal_attempts": 3,
  "last_wake_at": null,
  "wake_failure_count": 0,
  "next_wake_not_before": null,
  "started_at": null,
  "stopped_at": null,
  "created_at": "2026-05-12T...",
  "updated_at": "2026-05-12T..."
}

curl "https://api.veri.studio/v1/deployments?status=serving" \
  -H "Authorization: Bearer vk_..."

curl -X PATCH "https://api.veri.studio/v1/deployments/$DEPLOYMENT_ID" \
  -H "Authorization: Bearer vk_..." -H "Content-Type: application/json" \
  -d '{"scale_to_zero_window_seconds": null, "idle_delete_after_days": null}'

dep = client.deployments.update(deployment_id, min_replicas=1, max_replicas=4)
# Back to the defaults (sends an explicit null):
dep = client.deployments.update(
    deployment_id, clear_scale_to_zero_window_seconds=True, clear_idle_delete_after_days=True
)

curl "https://api.veri.studio/v1/deployments/$DEPLOYMENT_ID/replicas" \
  -H "Authorization: Bearer vk_..."

curl -X POST "https://api.veri.studio/v1/deployments/$DEPLOYMENT_ID/replicas/$REPLICA_ID/drain?replace=true" \
  -H "Authorization: Bearer vk_..."

client.deployments.wake(deployment_id)

final = client.deployments.stop(deployment_id)
print(final.status)

{
  "model": "my-deploy",
  "messages": [{"role": "user", "content": "Hello"}],
  "temperature": 0.7,
  "max_tokens": 256
}

{
  "id": "chatcmpl-abc",
  "object": "chat.completion",
  "created": 1715000000,
  "model": "my-deploy",
  "choices": [
    {"index": 0, "message": {"role": "assistant", "content": "..."}, "finish_reason": "stop"}
  ],
  "usage": {"prompt_tokens": 10, "completion_tokens": 5, "total_tokens": 15}
}

{
  "error": {
    "code": "deployment_waking",
    "message": "Deployment was scaled to zero and is waking up; retry shortly",
    "retry_after_seconds": 30
  }
}

response = client.deployments.chat(
    deployment_id,
    messages=[{"role": "user", "content": "Hi"}],
    model="my-deploy",                  # the deployment's name
    on_waking=lambda msg: print(msg),   # optional progress hook
    wake_timeout_s=1320,                # 0 = fail fast
    upstream_retries=3,                 # bounded retry on upstream_unavailable (0 = never)
    timeout_s=120,                      # HTTP timeout for this call (long non-streaming answers)
)

{
  "error": {
    "code": "upstream_unavailable",
    "message": "...",
    "retry_after_seconds": 5
  }
}

curl https://api.veri.studio/v1/deployments/$DEPLOYMENT_ID/responses \
  -H "Authorization: Bearer vk_..." \
  -H "Content-Type: application/json" \
  -d '{"model": "my-deploy", "input": "Hello", "max_output_tokens": 64}'

curl "https://api.veri.studio/v1/deployments/$DEPLOYMENT_ID/requests?limit=50" -H "Authorization: Bearer vk_..."

{
  "data": [
    {
      "object": "deployment_request",
      "id": "9b1f2c4a7d01",
      "deployment_id": "3f9c2a7d1b04",
      "model_version_id": "mv1a2b3c4d5e6f",
      "prompt_tokens": 12,
      "completion_tokens": 8,
      "latency_ms": 245.3,
      "status_code": 200,
      "error_message": null,
      "created_at": "2026-05-12T..."
    }
  ],
  "has_more": true
}

{
  "total_requests": 42,
  "total_prompt_tokens": 5230,
  "total_completion_tokens": 1840,
  "total_cost_usd": 2.6,
  "avg_latency_ms": 1240.5,
  "p50_latency_ms": 980.0,
  "p90_latency_ms": 2100.0,
  "p95_latency_ms": 2650.0,
  "p99_latency_ms": 4100.0,
  "error_rate": 0.024,
  "uptime_seconds": 3600.0
}

dep = client.deployments.create(model="...", source="huggingface", name="...", gpu={...})
dep.wait()                                          # polls until the status settles
response = dep.chat([{"role": "user", "content": "Hi"}])
metrics = dep.metrics()
recent = dep.requests(limit=20)
dep.stop()