-
Notifications
You must be signed in to change notification settings - Fork 217
Open
Description
The documentation currently does not explain how to set up GenAIScript with a AWS inference profile, which is required to access the more powerful models. Could you please provide guidance on how to properly configure GenAIScript in this scenario?
I attempted the following setup, but it failed:
DEBUG=* npx --yes genaiscript@latest run translate --model arn:aws:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0 --provider anthropic_bedrockgenaiscript:models resolving model for 'arn:aws:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0' +297ms
genaiscript:models candidate arn:aws:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0 +0ms
genaiscript:models resolving arn:aws:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0 +0ms
genaiscript:config:env parsing token for 'arn:aws:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0' +297ms
genaiscript:config:env looking for ARN_AWS_BEDROCK_US_EAST_1_024871859028_INFERENCE_PROFILE_GLOBAL_ANTHROPIC_CLAUDE_SONNET_4_20250514_V1_0_BEDROCK_US_EAST_1_024871859028_INFERENCE_PROFILE_GLOBAL_ANTHROPIC_CLAUDE_SONNET_4_20250514_V1_0_... +1ms
genaiscript:config:env looking for ARN_AWS_BEDROCK_US_EAST_1_024871859028_INFERENCE_PROFILE_GLOBAL_ANTHROPIC_CLAUDE_SONNET_4_20250514_V1_0_... +0ms
genaiscript:config:env looking for ARN_... +0ms
genaiscript:config:env looking for AWS_BEDROCK_US_EAST_1_024871859028_INFERENCE_PROFILE_GLOBAL_ANTHROPIC_CLAUDE_SONNET_4_20250514_V1_0_... +0ms
genaiscript:config:env no matching provider found, returning undefined +0ms
genaiscript:host:node no token found for arn:aws:bedrock:us-east-1:024871859028:inference-profile/global.anthropic.claude-sonnet-4-20250514-v1:0 +295ms
genaiscript:models configuration not found +1ms
genaiscript:error {
genaiscript:error name: 'Error',
genaiscript:error message: 'model configuration not found',
genaiscript:error stack: 'Error: model configuration not found\n' +
genaiscript:error ' at runPrompt (file:///Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@genaiscript/core/dist/esm/runpromptcontext.js:647:23)\n' +
genaiscript:error ' at process.processTicksAndRejections (node:internal/process/task_queues:105:5)\n' +
genaiscript:error ' at async <anonymous> (/Users/markus/src/TLA/_specs/TLAiBench/genaisrc/translate.genai.mts:71:24)'
genaiscript:error } +2ms
script {
script messages: [],
script text: '',
script reasoning: false,
script finishReason: 'fail',
script error: {
script name: 'Error',
script message: 'model configuration not found',
script stack: 'Error: model configuration not found\n' +
script ' at runPrompt (file:///Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@genaiscript/core/dist/esm/runpromptcontext.js:647:23)\n' +
script ' at process.processTicksAndRejections (node:internal/process/task_queues:105:5)\n' +
script ' at async <anonymous> (/Users/markus/src/TLA/_specs/TLAiBench/genaisrc/translate.genai.mts:71:24)'
script }
script } +0msFailure when trying without inference profile:
400 Invocation of model ID anthropic.claude-sonnet-4-20250514-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model.
Error: 400 Invocation of model ID anthropic.claude-sonnet-4-20250514-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model.
at APIError.generate (/Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@anthropic-ai/sdk/src/core/error.ts:63:14)
at AnthropicBedrock.makeStatusError (/Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@anthropic-ai/sdk/src/client.ts:410:28)
at AnthropicBedrock.makeRequest (/Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@anthropic-ai/sdk/src/client.ts:643:24)
at process.processTicksAndRejections (node:internal/process/task_queues:105:5)
status: 400
headers: {}
requestID: null
error:
message: Invocation of model ID anthropic.claude-sonnet-4-20250514-v1:0 with
on-demand throughput isn’t supported. Retry your request with the ID or ARN
of an inference profile that contains this model.
genaiscript:error {
genaiscript:error status: 400,
genaiscript:error headers: {},
genaiscript:error requestID: null,
genaiscript:error error: {
genaiscript:error message: 'Invocation of model ID anthropic.claude-sonnet-4-20250514-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model.'
genaiscript:error },
genaiscript:error name: 'Error',
genaiscript:error message: '400 Invocation of model ID anthropic.claude-sonnet-4-20250514-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model.',
genaiscript:error stack: 'Error: 400 Invocation of model ID anthropic.claude-sonnet-4-20250514-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model.\n' +
genaiscript:error ' at APIError.generate (/Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@anthropic-ai/sdk/src/core/error.ts:63:14)\n' +
genaiscript:error ' at AnthropicBedrock.makeStatusError (/Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@anthropic-ai/sdk/src/client.ts:410:28)\n' +
genaiscript:error ' at AnthropicBedrock.makeRequest (/Users/markus/.npm/_npx/4f85e8f904063418/node_modules/@anthropic-ai/sdk/src/client.ts:643:24)\n' +
genaiscript:error ' at process.processTicksAndRejections (node:internal/process/task_queues:105:5)'
genaiscript:error } +1ms
╰─🏁 anthropic_bedrock:anthropic.claude-sonnet-4-20250514-v1:0 ✉ 2 495ms ⇅ 0t ↑0t ↓0tMetadata
Metadata
Assignees
Labels
No labels