What causes "No generated images observed before timeout." in jackwener/OpenCLI?

Increase --timeout (e.g. 300-600) and retry — image generation often exceeds 180s under load

jackwener/OpenCLI · error · TimeoutError

qianwen image

qianwen image

Error message

No generated images observed before timeout.

What it means

clis/qwen/image.js polls waitForImageUrls(page, targetId, timeout) for generated image URLs; if the status comes back 'timeout' it throws TimeoutError('qianwen image', timeout, 'No generated images observed before timeout.'). Image generation simply did not produce any <img> result within the allotted seconds.

Source

Thrown at clis/qwen/image.js:152

        if (!send?.ok) {
            if (await hasLoginGate(page)) throw authRequired();
            throw new CommandExecutionError(send?.reason || 'Failed to send Qianwen image prompt');
        }

        // Grab the newest assistant bubble id after send by polling briefly
        let targetId = '';
        for (let i = 0; i < 5; i += 1) {
            await page.wait(1);
            const bubbles = await getMessageBubbles(page);
            const lastAnswer = [...bubbles].reverse().find((b) => b.role === 'Assistant');
            if (lastAnswer) { targetId = lastAnswer.id; break; }
        }

        const waitResult = await waitForImageUrls(page, targetId, timeout);
        const link = await page.evaluate('window.location.href').catch(() => 'https://www.qianwen.com/');
        if (waitResult.status === 'auth_required') throw authRequired();
        if (waitResult.status === 'timeout') {
            throw new TimeoutError('qianwen image', timeout, 'No generated images observed before timeout.');
        }

        const urls = waitResult.urls;
        if (skipDownload) {
            return [{ Status: '🎨 generated', File: null, Link: link }];
        }

        const stamp = Date.now();
        const results = [];
        for (let i = 0; i < urls.length; i += 1) {
            const url = urls[i];
            const asset = await fetchImageAsset(page, url);
            if (!asset?.ok) {
                throw new CommandExecutionError(`Failed to fetch generated Qianwen image ${i + 1}: status=${asset?.status || '?'}`);
            }
            const suffix = urls.length > 1 ? `_${i + 1}` : '';
            const ext = extFromMime(asset.mime);
            const filePath = path.join(outputDir, `qianwen_${stamp}${suffix}${ext}`);

View on GitHub (pinned to 49907e53dc)

Solutions

Increase --timeout (e.g. 300-600) and retry — image generation often exceeds 180s under load
Simplify or rephrase the prompt; content-policy refusals produce no images at all
Verify the sent prompt actually started generating (check the chat in the browser)
Retry later if the service is degraded/queued — server-side latency is outside your control

Example fix

// before
clis qwen image --prompt 'elaborate fantasy scene' --timeout 60
// after
clis qwen image --prompt 'elaborate fantasy scene' --timeout 600

Defensive patterns

Strategy: retry

Validate before calling

null

Type guard

null

Try / catch

try {
  const results = await runQianwenImage(prompt, { timeout: 600 });
} catch (e) {
  if (e instanceof TimeoutError && e.operation === 'qianwen image') {
    console.warn(`No images after ${e.timeout}s; retrying once with a simpler prompt`);
    await runQianwenImage(simplify(prompt), { timeout: 600 });
  } else throw e;
}

Prevention

Use generous timeouts (300-600s) for image generation, especially under load
Avoid prompts likely to trigger content refusals, which yield no images
Check the chat in the browser to confirm generation actually starts
Treat repeated timeouts as service degradation and back off rather than hammering retries

When it happens

Trigger: Image generation takes longer than the --timeout value (default 180s); the prompt was rejected or the model returned text only; generation stalled after send; the target assistant bubble (targetId) never received image elements.

Common situations: Complex prompts queuing behind heavy server load; free-tier rate limiting slowing generation; very short custom timeouts (e.g. --timeout 30) on a feature that usually takes 1-3 minutes; content-policy refusal producing no images.

Understand the failure class

Timeouts: ETIMEDOUT, deadlines, and hung requests — what actually expires when a request times out.

Related errors

AI-assisted analysis of jackwener/OpenCLI@49907e53dc (2026-08-29). Data as JSON: /api/errors/e07d13db4093ea7e. Report an issue: GitHub.