conductor-oss/conductor · error · UnsupportedOperationException

Image generation not supported by the model yet

Error message

Image generation not supported by the model yet

What it means

Bedrock.getImageModel() throws unconditionally because the Bedrock provider implementation does not wire up an image generation model. While AWS Bedrock does host image-generation models (Stability AI, Amazon Titan Image Generator), this Conductor provider only implements chat (via BedrockProxyChatModel) and embeddings (via Cohere). The interface contract is fulfilled by throwing UnsupportedOperationException.

Source

Thrown at ai/src/main/java/org/conductoross/conductor/ai/providers/bedrock/Bedrock.java:148

        String model = input.getModel();
        log.info("\n\nusing bedrock model: {}", model);
        return BedrockChatOptions.builder()
                .model(model)
                .maxTokens(input.getMaxTokens())
                .topP(input.getTopP())
                .temperature(input.getTemperature())
                .toolCallbacks(getToolCallback(input))
                .stopSequences(input.getStopWords())
                .frequencyPenalty(input.getFrequencyPenalty())
                .topK(input.getTopK())
                .internalToolExecutionEnabled(false)
                .presencePenalty(input.getPresencePenalty())
                .build();
    }

    @Override
    public ImageModel getImageModel() {
        throw new UnsupportedOperationException("Image generation not supported by the model yet");
    }

    private Map<String, Object> getEmbeddingRequest(String modelId, String text) {
        return Map.of(
                "input_type", "search_document",
                "embedding_types", List.of("float"),
                "texts", List.of(text));
    }

    @SneakyThrows
    private List<Float> extractEmbeddings(String modelId, InvokeModelResponse response) {
        if (!modelId.startsWith("cohere.")) {
            throw new RuntimeException("Unsupported model " + modelId);
        }
        byte[] byteArray = response.body().asByteArray();
        Map<String, Map<String, Object>> ressMap = om.readValue(byteArray, Map.class);
        List<List<Float>> floats = (List<List<Float>>) ressMap.get("embeddings").get("float");
        return floats.get(0);

View on GitHub (pinned to cf7c3e4a8a)

Solutions

  1. Route image generation to a provider with an implemented ImageModel: 'openai', 'azure_openai', 'vertex_ai'/'google_gemini', or 'stabilityai'.
  2. If you need Bedrock image models specifically, implement getImageModel() in the Bedrock provider using an InvokeModel call to stability.stable-image-* or amazon.titan-image-*.
  3. Check provider capabilities before dispatching.

Example fix

// before
String provider = "bedrock";
AIModel model = providerRegistry.get(provider);
ImageModel img = model.getImageModel(); // throws

// after
String provider = "openai";
AIModel model = providerRegistry.get(provider);
ImageModel img = model.getImageModel();
Defensive patterns

Strategy: validation

Validate before calling

// Validate Bedrock supports image generation before calling
private static final Set<String> IMAGE_CAPABLE_PROVIDERS =
    Set.of("openai", "azure_openai", "vertex_ai", "google_gemini", "stabilityai");

void validateImageCapability(String providerName) {
    if (!IMAGE_CAPABLE_PROVIDERS.contains(providerName)) {
        throw new IllegalArgumentException(
            "Provider '" + providerName + "' does not support image generation");
    }
}

Type guard

static boolean supportsImageGeneration(AIModel model) {
    try {
        model.getImageModel();
        return true;
    } catch (UnsupportedOperationException e) {
        return false;
    }
}

Try / catch

try {
    ImageModel img = bedrock.getImageModel();
} catch (UnsupportedOperationException e) {
    throw new IllegalArgumentException(
        "Bedrock does not implement image generation in this provider. "
        + "Use openai, vertex_ai, or stabilityai.", e);
}

Prevention

When it happens

Trigger: Calling getImageModel() on a Bedrock provider instance — e.g. a GenerateImage task routed to provider 'bedrock'.

Common situations: Routing an image generation task to Bedrock expecting it to use Stability/Titan image models. A generic dispatcher that assumes all providers support image generation.

Related errors

AI-assisted analysis of conductor-oss/conductor@cf7c3e4a8a (2026-08-14). Data as JSON: /api/errors/bc787facc0caa388. Report an issue: GitHub.