CherryHQ/cherry-studio · error · Error

errorData.error?.message || 'Image generation failed'

Error message

errorData.error?.message || 'Image generation failed'

What it means

Thrown by OvmsTransport.submit() when POST {baseURL}/images/generations returns non-ok. The message comes from the JSON body's nested error.message (OVMS uses { error: { message } }), falling back to 'Image generation failed'; the .catch produces `HTTP {status}` when the body is not JSON. OVMS runs without auth, so a non-ok is a server/model problem, not auth.

Source

Thrown at src/main/ai/provider/custom/ovms/ovmsTransport.ts:60

    // stripped by passthroughExtras). Native size/seed come from `input.*`.
    const requestBody = {
      model: input.modelId,
      prompt: input.prompt ?? '',
      size: input.size ?? '512x512',
      num_inference_steps: typeof bag.num_inference_steps === 'number' ? bag.num_inference_steps : 4,
      rng_seed: input.seed ?? 0
    }

    const response = await fetch(`${this.baseURL}/images/generations`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify(requestBody),
      signal: input.signal
    })

    if (!response.ok) {
      const errorData = await response.json().catch(() => ({ error: { message: `HTTP ${response.status}` } }))
      throw new Error(errorData.error?.message || 'Image generation failed')
    }

    const data = await response.json()
    const items = Array.isArray(data?.data) ? data.data : []

    const base64s = items
      .filter((item: { b64_json?: string }) => item.b64_json)
      .map((item: { b64_json: string }) => `data:image/png;base64,${item.b64_json}`)
    if (base64s.length > 0) {
      return { imageUrls: base64s }
    }

    const urls = items.filter((item: { url?: string }) => item.url).map((item: { url: string }) => item.url)
    return { imageUrls: urls }
  }
}

export function createOvmsTransport(settings: OvmsTransportSettings): OvmsTransport {

View on GitHub (pinned to 726446b54c)

Solutions

  1. Confirm OVMS is running and the diffusion model is deployed (check OVMS /models endpoint or logs).
  2. Verify settings.imageBaseURL / DEFAULT_OVMS_BASE_URL (http://localhost:8000) actually serves /images/generations.
  3. Read the embedded error.message — OVMS usually names the missing model or the invalid parameter.
  4. Validate requestBody fields (size in WxH, num_inference_steps within model limits) before submit.

Example fix

// before
throw new Error(errorData.error?.message || 'Image generation failed')
// after — include status and host for diagnostics
throw new Error(`OVMS (${this.baseURL}) image generation failed (${response.status}): ${errorData.error?.message || response.statusText}`)
Defensive patterns

Strategy: validation

Validate before calling

// Confirm OVMS is reachable and the model is deployed
const res = await fetch(`${imageBaseURL}/models`)
if (!res.ok) throw new Error(`OVMS not reachable at ${imageBaseURL}`)

Type guard

export function isOvmsImageError(e: unknown): boolean {
  return e instanceof Error && /Image generation failed/.test(e.message)
}

Try / catch

try {
  await transport.submit(input)
} catch (e) {
  if (e instanceof Error) {
    // surface the embedded error.message (OVMS names the missing model / bad param)
    throw new Error(`OVMS image generation failed: ${e.message}`)
  }
  throw e
}

Prevention

When it happens

Trigger: The OVMS OpenVINO Model Server returns non-2xx from /images/generations. Concrete causes: the diffusion model is not loaded/deployed, OVMS not running on the configured imageBaseURL, malformed body (size/steps out of range), model crash, or wrong endpoint path.

Common situations: OVMS started without the image model deployed, imageBaseURL pointing at the chat host (which has no /images/generations), model cold-load failure, or stale OVMS version.

Related errors

AI-assisted analysis of CherryHQ/cherry-studio@726446b54c (2026-08-12). Data as JSON: /api/errors/d7578261f86613a1. Report an issue: GitHub.